<%BANNER%>

Optional Subject-Verb Agreement in Persian

MISSING IMAGE

Material Information

Title:
Optional Subject-Verb Agreement in Persian
Physical Description:
1 online resource (431 p.)
Language:
english
Creator:
Feizmohammadpour, Aazamosadat
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Linguistics
Committee Chair:
Cowles, Heidi Wind
Committee Members:
Henderson, Brent Mykel
Kaan, Edith
Altmann, Lori J

Subjects

Subjects / Keywords:
agreement -- persian
Linguistics -- Dissertations, Academic -- UF
Genre:
Linguistics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Persian exhibits subject-verb number agreement but agreement is optional if the subject of the sentence is an inanimate plural noun. The goal of this dissertation was to examine the effect of grammatical and conceptual factors on the production of optional number agreement in Persian. Some additional features of Persian make it an especially interesting language to study agreement phenomena in. First, while most work on agreement has dealt with ungrammatical sentences, Persian allows the grammatical use of either number matched or mismatched verbs for inanimate plural noun subjects. Second, SOV word order of its sentences creates the opportunity to test the attraction effect of an intermediary object. Third, Persian has two bound plural morphemes, /-ha/ and /-an/,suitable to test the effect of morphology. Finally, using the compound verbs in Persian provides this opportunity to restrict the participants to complete the preambles with the desired verb (verbal constituent), keeping the effect of verb type constant. A judgment task tested the participants’ acceptance of sentences in Experiment 1 with matched and mismatched verbs and their interpretation of the thematic role of the subject nouns. The results showed a greater tendency to accept mismatched verbs in sentences with “patient” and “patient-like” subjects. Experiment 2tested the effect of verb type and verb tense. The results were consistent with Experiment 1; participants produced more mismatched verbs with “patient” and “patient-like”subjects. Experiment 3 tested the effect of unity vs. individuality conceptualization of subject noun. The result showed more mismatched verbs produced with subjects conceptualized as a unit rather than as individuals. Experiment4 tested the effect of concreteness of subject noun and the result showed more mismatched verbs produced with abstract nouns rather than concrete nouns. Finally, Experiment 5 showed the attraction effect of intermediary object nouns.The results showed an effect of number and morphology features of object nouns in SOV sentences. The results of all the experiments show the effect of conceptual information in the subject-verb agreement. These results demonstrate the importance of non-syntactic factors in subject-verb number agreement in Persian.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Aazamosadat Feizmohammadpour.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Cowles, Heidi Wind.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045325:00001

MISSING IMAGE

Material Information

Title:
Optional Subject-Verb Agreement in Persian
Physical Description:
1 online resource (431 p.)
Language:
english
Creator:
Feizmohammadpour, Aazamosadat
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Linguistics
Committee Chair:
Cowles, Heidi Wind
Committee Members:
Henderson, Brent Mykel
Kaan, Edith
Altmann, Lori J

Subjects

Subjects / Keywords:
agreement -- persian
Linguistics -- Dissertations, Academic -- UF
Genre:
Linguistics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Persian exhibits subject-verb number agreement but agreement is optional if the subject of the sentence is an inanimate plural noun. The goal of this dissertation was to examine the effect of grammatical and conceptual factors on the production of optional number agreement in Persian. Some additional features of Persian make it an especially interesting language to study agreement phenomena in. First, while most work on agreement has dealt with ungrammatical sentences, Persian allows the grammatical use of either number matched or mismatched verbs for inanimate plural noun subjects. Second, SOV word order of its sentences creates the opportunity to test the attraction effect of an intermediary object. Third, Persian has two bound plural morphemes, /-ha/ and /-an/,suitable to test the effect of morphology. Finally, using the compound verbs in Persian provides this opportunity to restrict the participants to complete the preambles with the desired verb (verbal constituent), keeping the effect of verb type constant. A judgment task tested the participants’ acceptance of sentences in Experiment 1 with matched and mismatched verbs and their interpretation of the thematic role of the subject nouns. The results showed a greater tendency to accept mismatched verbs in sentences with “patient” and “patient-like” subjects. Experiment 2tested the effect of verb type and verb tense. The results were consistent with Experiment 1; participants produced more mismatched verbs with “patient” and “patient-like”subjects. Experiment 3 tested the effect of unity vs. individuality conceptualization of subject noun. The result showed more mismatched verbs produced with subjects conceptualized as a unit rather than as individuals. Experiment4 tested the effect of concreteness of subject noun and the result showed more mismatched verbs produced with abstract nouns rather than concrete nouns. Finally, Experiment 5 showed the attraction effect of intermediary object nouns.The results showed an effect of number and morphology features of object nouns in SOV sentences. The results of all the experiments show the effect of conceptual information in the subject-verb agreement. These results demonstrate the importance of non-syntactic factors in subject-verb number agreement in Persian.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Aazamosadat Feizmohammadpour.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Cowles, Heidi Wind.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045325:00001


This item has the following downloads:


Full Text

PAGE 1

1 OPTIONAL SUBJECT VERB AGREEMENT IN PERSIAN By AAZAMOSADAT FEIZMOHAMMADPOUR A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR O F PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Aazamosadat Feizmohammadpour

PAGE 3

3 To : Iran and Persian langu a ge

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank the committee members, Lori Altman, Edith Kaan, Brent H enderson for all of their interest, guidance, and insightful comments throughout the process of this research. My special thanks go to my advisor, Wind Cowles, whose support, patience, guidance and advice over the years have been invaluable to me. I would also like to thank Arnout Koornneef for his kind assistance and expertise at many stages of the data analysis. Still more thanks to: my Persian linguist ic s professors whose scientific works enlighten ed my path during this dissertation and inspired me to st udy Linguistics and do research, the professors and staff of the L inguistics D epartment at the University of Florida for all their kind assistance and support, Tiffany Judy and Sara Court for their great support and assistance in proof readin g chapters of the dissertation, Iranian community at Gainesville who voluntarily participated in the ex periments of this dissertation, and Sabrina Majlesara for he r continuous encouragement and positive energy My sincere thanks go to my parents for all their hard effor ts to support me throughout my life and education without them I could not be here. Finally, I would also like to thank my husband, Heshmat Sarouei for his patience, encouragement and help during the years of my study His exceptional supportive assistan ce in taking care of our baby Niki during the final stages of dissertation enabled me to write the final chapters and get prepared for the defens e, without his help I could not have finish ed th is work.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 16 Overview ................................ ................................ ................................ ................. 16 Some Characterist ics of Persian ................................ ................................ ............. 18 Singular/ Plural Nouns ................................ ................................ ...................... 19 Simple and Compound Verbs in Persian ................................ .......................... 22 Combination ................................ ................................ ............................... 24 Incorporation ................................ ................................ .............................. 26 Passive and Inchoative in Persian ................................ ................................ .... 30 Passive ................................ ................................ ................................ ...... 30 Inchoative ................................ ................................ ................................ ... 31 Subject Verb Agreement in Persian ................................ ................................ 32 Probable Factors In fluencing Agreement in Persian ................................ ........ 36 The Goal and Outline of the Dissertation ................................ ................................ 41 The Importance of the Study ................................ ................................ ................... 43 2 REVIEW OF RELATED LITERATURE ................................ ................................ ... 47 Overview ................................ ................................ ................................ ................. 47 Psycholinguistic Accounts of Subject Verb Agreement Processing ........................ 48 Evidence for Minimalism ................................ ................................ ................... 50 Evidence for Maximalism ................................ ................................ .................. 53 Current Psycholinguistic Accounts and Theories ................................ .................... 56 Psycholinguistic Approaches and Agreement in Persian ................................ ........ 60 Syntactic Factors Influencing Subject Verb Production and Processing ................. 62 Attraction Effects ................................ ................................ .............................. 62 Only Syntactic Factors ................................ ................................ ...................... 64 Hierarchical Structure ................................ ................................ ....................... 68 Non Syntactic Factors influencing Subject verb Production and Processing .......... 70 Distributivity ................................ ................................ ................................ ...... 70 Concreteness/ Imageability of Distributive Nouns ................................ ............ 73

PAGE 6

6 Colle ctivity ................................ ................................ ................................ ........ 76 Gender Agreement (Conceptual Information) ................................ ................... 77 Morphology ................................ ................................ ................................ ....... 78 Plausibility ................................ ................................ ................................ ........ 79 Comprehension of Agreement ................................ ................................ ................ 80 3 EXPERIMENT 1 JUDGMENT TASK ................................ ................................ ..... 87 Overview ................................ ................................ ................................ ................. 87 Goals ................................ ................................ ................................ ...................... 88 Acceptability Goals ................................ ................................ ........................... 88 Role Interpretation Goals ................................ ................................ .................. 89 Method ................................ ................................ ................................ .................... 90 Participants ................................ ................................ ................................ ....... 90 Materials ................................ ................................ ................................ ........... 90 Past and present tense in Persian ................................ ............................. 91 Verb types ................................ ................................ ................................ .. 95 Group A Unaccusative verbs ................................ ................................ ... 95 Group B Entity specific inchoative verbs ................................ ................. 97 Group C Verbs of emission ................................ ................................ ...... 98 Group D Verbs of instrument ................................ ................................ ... 98 Group E Verbs of agency ................................ ................................ ....... 100 Group F Passive verbs ................................ ................................ .......... 101 Group G ................................ ............................ 101 Group H Inchoat ................................ .............. 102 Procedure ................................ ................................ ................................ ....... 104 Design and Data Analysis ................................ ................................ .............. 105 Acceptability ................................ ................................ ............................. 105 Role interpretation ................................ ................................ .................... 106 Questions of the Experiment ................................ ................................ .......... 107 Acceptability ................................ ................................ ............................. 107 Role interpretation ................................ ................................ .................... 107 Predictions ................................ ................................ ................................ ...... 108 Acceptability ................................ ................................ ............................. 108 Role interpretation ................................ ................................ .................... 108 Results ................................ ................................ ................................ ........... 109 Result of acceptability. ................................ ................................ ............. 109 Result of role interpretation. ................................ ................................ ..... 112 Discussion ................................ ................................ ................................ ............ 115 Conclusion ................................ ................................ ................................ ............ 118 4 EXPERIMENT 2 EFFECT OF VERB TYPE AND TENSE ................................ ... 135 Overview ................................ ................................ ................................ ............... 135 Method ................................ ................................ ................................ .................. 136 Participants ................................ ................................ ................................ ..... 136 Materials ................................ ................................ ................................ ......... 136

PAGE 7

7 Procedure ................................ ................................ ................................ ....... 138 Design and Data Analysis ................................ ................................ .............. 139 Predictions ................................ ................................ ................................ ...... 140 Results ................................ ................................ ................................ ........... 140 Discussion ................................ ................................ ................................ ............ 142 Conclusion ................................ ................................ ................................ ............ 144 5 EXPERIMENT 3 EFFECT OF UNITY OF SUBJECT NOUN ............................... 153 Overview ................................ ................................ ................................ ............... 153 Method ................................ ................................ ................................ .................. 155 Participants ................................ ................................ ................................ ..... 155 Materials ................................ ................................ ................................ ......... 156 Procedure ................................ ................................ ................................ ....... 159 Design and Data Analysis ................................ ................................ .............. 160 Results ................................ ................................ ................................ ........... 161 Discussion ................................ ................................ ................................ ............ 163 Conclusion ................................ ................................ ................................ ............ 165 6 EXPERIMENT 4 EFFECT OF SUBJECT NOUN CONCRETENESS .................. 174 Overview ................................ ................................ ................................ ............... 174 Method ................................ ................................ ................................ .................. 175 Participants ................................ ................................ ................................ ..... 175 Materials ................................ ................................ ................................ ......... 175 Procedure ................................ ................................ ................................ ....... 177 Design and Data Analysis ................................ ................................ .............. 178 Results ................................ ................................ ................................ ........... 179 Discuss ion ................................ ................................ ................................ ............ 180 Conclusion ................................ ................................ ................................ ............ 181 7 EXPERIMENT 5 OBJECT ATTRACTION EFFECT ................................ ............. 190 Overview ................................ ................................ ................................ ............... 1 90 Methods ................................ ................................ ................................ ................ 193 Participants ................................ ................................ ................................ ..... 193 Materials ................................ ................................ ................................ ......... 194 Procedure ................................ ................................ ................................ ....... 199 Design and Data Analysis ................................ ................................ .............. 200 Resul ts ................................ ................................ ................................ ........... 201 Discussion ................................ ................................ ................................ ............ 203 Conclusion ................................ ................................ ................................ ............ 206 8 CONCLUSION AND DISCUSSION ................................ ................................ ...... 215 Ove rview ................................ ................................ ................................ ............... 215 Experiment 1 ................................ ................................ ................................ ......... 217 Experiment 2 ................................ ................................ ................................ ......... 220

PAGE 8

8 Experiment 3 ................................ ................................ ................................ ......... 224 Experiment 4 ................................ ................................ ................................ ......... 226 Experiment 5 ................................ ................................ ................................ ......... 227 Suggestions for Further Study ................................ ................................ .............. 233 APPENDIX A GLOSS JUDGMENT TASK ................................ ................................ ................. 235 B GLOSS VERB TYPE VERB TENSE EXPERIMENT ................................ ........... 312 C GLOSS UNITY EXPERIMENT ................................ ................................ ............ 346 D GLOSS CONCRETENESS EXPERIMENT ................................ ......................... 367 E GLOSS ATTRACTION EXPER IMENT ................................ ................................ 375 LIST OF REFERENCES ................................ ................................ ............................. 423 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 431

PAGE 9

9 LIST OF TABLES Table page 3 1 Acceptability, test of within su bject effects (per participants) ............................ 120 3 2 Acceptability, test of within subject effects (per items) ................................ ...... 121 3 3 Mean proportion of unacceptability of each Verb Type ................................ ..... 122 3 4 Pairwise comparison of the mean prop ortion of unacceptability of mismatched verb for each Verb Type ................................ ............................... 123 3 5 Role interpretation, test of within subject effects (per participant) ..................... 125 3 6 Role interpretation, test of within subject effects (per items) ............................ 126 3 7 Mean proportion of patient role interpretation for different Verb Type .............. 127 3 8 Pairwise comparison of the mean proportion of acted on interpretation of the subject nouns for each Verb Type ................................ ................................ .... 128 3 9 Mean proportion of patient role interpretation for different Verb Tense ............ 130 3 10 Pairwise comparison of patient interpretation for different Verb Tense ............ 131 3 11 Mean proportion of patient interpretation for different verb number .................. 132 3 12 Pairwise comparison of patient interpretation for different verb number ........... 133 3 13 Pat ient interpretation for subjects (interaction of verb tense and verb number) 134 4 1 Sample preamble of verb type verb tens e experiment & possible response .... 146 4 2 Test of within subject effect (per participants) ................................ ................... 148 4 3 Test of within subject effect (per items) ................................ ............................ 149 4 4 Mean proportion of SG verbs produced in present and past sentences ........... 150 4 5 Mean proportion of SG verbs produced in different verb groups ...................... 151 4 6 Effect of interaction of Verb Type and Verb Tense in the production of mismatched verbs ................................ ................................ ............................ 152 5 1 Conditions of subject phrases ................................ ................................ ........... 167 5 2 Sa mple preamble and possible target response for each condition ................. 168

PAGE 10

10 5 3 Test of within subject effect per participants ................................ ..................... 169 5 4 Test of within subject effect (per items) ................................ ............................ 170 5 5 Mean proportion of mismatched verb usage with different modifying prepositional phrase ................................ ................................ ......................... 171 5 6 Pairwise comparisons of proportion of mismatched verb usage with different modifying prepositional phrase ................................ ................................ ......... 172 5 7 Mean proportion of singular verb usage (interaction of prepositional phrase and structure ................................ ................................ ................................ ..... 173 6 1 Sample of preamble and possible response for concreteness experiment ....... 182 6 2 Test of within subject effect (per participants) ................................ ................... 183 6 3 Test of within subject effect (per items) ................................ ............................ 184 6 4 Mean proportion of mismatched verb with concrete and abs tract subject ........ 185 6 5 Pairwise comparison of concrete vs. abstract subject ................................ ...... 186 6 6 Mean proportion of singular verb produced in different structure ...................... 187 6 7 Pairwise comparison of inchoative and passive structures ............................... 188 6 8 Mean proportion of singular verb usage (interaction of concreteness and structure) ................................ ................................ ................................ .......... 189 7 1 Sample preamble and possible response for each condition ........................... 208 7 2 Test of within subject effect per participants (conditions: APH, ASO, IPH, ISO) ................................ ................................ ................................ .................. 209 7 3 Test of within subject effect per items (conditions: APH, ASO, IPH, ISO) ........ 210 7 4 Test of within subject effect per participants (conditions: APN, ASO, IPH, I SO) ................................ ................................ ................................ .................. 211 7 5 Test of within subject effect per items (conditions: APN, ASO, IPH, ISO) ........ 212 7 6 Mean proportion of singular verb usage with different verb groups .................. 213 7 7 Pairwise comparison of plural morphemes / ha/ and / an/ ................................ 214 A 1 List of the stimuli for Judgment experiment (condition: Past Singular) ............. 235 A 2 List of the stimuli for judgment experiment (Condition: Past Plural) .................. 249

PAGE 11

11 A 3 List of the stimuli for judgment experiment (Condition: Present S ingular) ........ 264 A 4 List of the stimuli for judgment experiment (Condition: Present Plural) ............ 278 A 5 List of the filler stimuli for judgment experiment (Condition: past) ..................... 292 A 6 List of the filler stimuli for judgment experiment (Condition: present) ............... 302 B 1 List of the target stimuli for Verb Type Verb Tense experiment ........................ 313 B 2 List of the filler items for Verb Type Verb Tense experiment ............................ 332 C 1 List of the stimuli for Unity experiment ................................ .............................. 347 C 2 List of the filler items for Unity & Concreteness experiment ............................. 359 D 1 List of the stimuli for Concreteness experiment ................................ ................ 367 E 1 List of the stimuli for Attraction experiment ................................ ....................... 376 E 2 List of the filler items for Attraction experiment ................................ ................. 416

PAGE 12

12 LIST OF FIGURES Figure page 2 1 Hierarchical encapsulated model by Bock and Levelt (1994) ............................. 84 2 2 Two accounts of the relationship between numerosity ................................ ....... 85 2 3 Marking and Morphing model of agreement for verb agreement and attraction 86

PAGE 13

13 LIST OF ABBREVIATIONS 3 S G THIRD PERSON SINGUAL R 3 PL THIRD PERSOAN PLULAR A CC ACCUSATIVE ANI ANIMATE INANI INANIMATE INCH INCHOATIVE IND INDIVIDUALIEZED N NOUN NO NUMBER NP NOUN PHRASE OM OBJECT MARKER PASS PASSIVE PRES PRESENT PL PLURAL PP PAST PARTICIPLE SG SINGUALR STRUC STRUCT URE VT VERB TYPE IP INDIVIDUALIZED PLURA L UP UNIFIED PLULAR US UNIFIED SINGULAR

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor o f Philosophy OPTIONAL SUBJECT VERB AGREEMENT IN PERSIAN By Aazamosadat Feizmohammadpour August 2013 Chair: Heidi Wind Cowles Major: Linguistics Persian exhibits subject verb number agreement but agreement is optional if the subject of the sentence is a n inanimate plural noun. The goal of this dissertation was to examine the effect of grammatical and conceptual factors on the production of optional number agreement in Persian. Some additional features of Persian make it an especially interesting language to study agreement phenomena in. First, while most work on agreement has dealt with ungrammatical sentences, Persian allows the grammatical use of either number matched or mismatched verb s for inanimate plural noun subjects. Second, SOV word order of its sentences creates the opportunity to test the attraction effect of an intermediary object. Third, Persian has two bound plural morphemes, / ha/ and / an/ suitable to test the effect of morphology. Finally, using the compound verbs in Persian provides this opportunity to restrict the participants to complete the preambles with the desired verb (verbal constituent), keeping the effect of verb type constant. nce of sentences in Experiment 1 with matched and mis matched verbs and their interpretation of the thematic role of the subject nouns The results showed a greater tendency to accept mismatched verbs in

PAGE 15

15 verb type and verb tense. The results were consistent with E xperiment 1 ; participants produce d more mismatched verbs with s Experiment 3 tested the effect of unity vs. individuality conceptualization of subject noun. The result showed mo re mismatched verbs produced with s ubjects conceptualized as a unit rather than as individual s Experiment 4 tested the effect of concreteness of subject noun and the result showed more mismatched verbs produced with abstract nouns rather than concrete nou ns. Finally, Experiment 5 showed the attraction effect of intermediary object noun s The results showed an effect of number and morphology features of object nouns in SOV sentences. The results of all the experiments show the effect of conceptual informati on in the subject verb agreement. These result s demonstrate the importance of non syntactic factors in subject verb number agreement in Persian.

PAGE 16

16 CHAPTER 1 INTRODUCTION Overview agreement of one kind or another, i n which the features of one word influence the form of another, syntactically related word. For example, English has subject verb number agreement: Main verbs in sentences change form depending on the plurality of the subject (although it is not overtly m arked present tense, as shown below in (1.1 and 1.2). (1.1) The boy is a student. // The boys are students. (1.2) The boy runs. // The boys run. In English, as in most languages with subject verb agreement, when agreement is not grammatical in standard dialects of English. While number agreement may seem to be somewhat esoteric, it has been of great inte rest to linguists and especially psycholinguists because agreement phenomena are quite common and can provide important evidence about how language is structured and how that structure is treated during language production and processing. Over the past 15 years, two major approaches to agreement processing have emerg ed, which may be characterized as the Minimalist Approach, in which agreement is governed solely by syntactic information, and the Maximalist Approach, in which both conceptual and syntactic inf ormation influence agreement during grammatical encoding. There is empirical evidence for both approaches, however, the

PAGE 17

17 data have always come from languages in which subject verb agreement is required, and so theories have necessarily relied on data from s tructures that are ungrammatical. In this dissertation, I am going to study these phenomena in Persian. The focus of my dissertation research is on the morphological agreement between subject nouns and verbs. Persian, too, has subject verb number agreement However, in Persian there is an interesting twist: When a plural subject is inanimate, number agreement is optional : the verb can take singular or plural form. Both of these forms are grammatical and acceptable both in spoken and written language. This i s illustrated in (1.3) below. (1.3) barg ha: mi riz e (cf. barg ha: mi riz n) leaf PL IND fall 3SG (cf. leaf PL IND fall 3PL) s Optional subject verb number agreement in Persian in which grammatical subject nouns allows us to study agreement using grammatically acceptable materials without requiring speakers to respond to or produce ungrammatical language. In addition, Persian has a nu mber of other features that make it an especially interesting language to study agreement phenomena in. First, it has canonical Subject Object Verb (SOV) word order which creates the opportunity to test the attraction effect on subject verb agreement when there is an intervening object noun. Second, Persian has two bound plural nominal morphemes, / ha/ and / an/, which creates an opportunity to test the effect of morphology. Finally, Persian has compound verbs, which allows us to provide constraining experi mental contexts for sentence production. These features, along with other important aspects of Persian will be discussed later in this chapter.

PAGE 18

18 The outline of this dissertation as follows. This chapter, Chapter 1, introduces the main questions and issues of this dissertation, and starts by providing the necessary background on Persian. It concludes with a summary of the main questions that this dissertation seeks to answer. Chapter 2 provides a review of the literature relevant to the processing of subjec t verb agreement. Chapters 3 7 discuss the five experiments th at I conducted to test the effects of different factors in optional agreement in Persian. Finally, Chapter 8 provides a summary and discussion of the results as a whole, and ends with some concl uding remarks. Some Characteristics of Persian In this section, I will provide information on the aspects of Persian that are most relevant for this dissertation. This section is not intended as a definitive grammar of Persian, but serves to lay the founda tion for the research questions outlined at the end of this chapter. Persian (also known as New Persian or Farsi) is an Indo European language that is the official language of Iran, with dialects also spoken in Afghanistan, Tajikistan, and Uzbekistan. The history of this language begins with Proto Iranian (1500 BC) and extends into Modern Persian, which is itself divided into three stages: early (800 AD 900 AD), classical (900 AD to 1800 AD), and contemporary Persian (1800 AD to present). Persian uses Ara bic script, with a few additional characters to accommodate Persian sounds (/p/, /zh/, /ch/, /g/) not present in Arabic. The unmarked word order of Persian is Subject Object Verb (SOV), however Persian allows scrambling in which many different word orders are permitted and used. It is a pro drop language (the subject can be deleted) and the verb ending agrees with the subject of the sentence. There are no overt case markers in Persian but the

PAGE 19

19 postposition /ra/, /ro/, or /o/ is the definite object marker. G ender is not morphologically reflected in this language: verbs, nouns, adjectives, and even pronouns are free from gender. Singular/ Plural Nouns Like English, nouns in Persian can be either singular (SG; referring to one entity) or plural (PL; referring t o more than one entity). The singular noun is the unmarked form as it does not have any overt morphological marker. Plurality is shown either by the number before the singular noun or by one of the plural markers suffixed to the noun. (1.4) gol yek gol flower one flower do gol gol ha two flower flower PL As can be seen in (1.4), if the number is explicitly given, then no plural morpheme is required (e.g. do gol ). For bare nouns, or nouns that do not have a numeric quantifier, a plural suffix must be added to the noun. The identity of this suffix is contingent on the animacy of the noun. The suffix / an/ is restricted to use with animate nouns and a small set of living inanimate t hings (e.g. humans, some animals, trees and plants) and any adjectives that are used as nouns. This plural marker, shown in (1.5), is mostly used in formal and written language. Normally, it is not used for borrowed words.

PAGE 20

20 (1.5) derxt an xub an dox tr an tree PL good PL girl PL zxmi yan geda yan prnde gan injured PL poor PL bird PL shir an *medad an *divar an lion PL penci l PL wall PL *miz an *rud an *xane gan desk PL river PL house PL An alternative to / an/ is the suffix / ha/. This suffix (whose surface form is sometimes / a/) is the unmarked plur al marker that is used for all animate and inanimate, Persian or borrowed nouns and adjectives used as nouns. It is used in both written and spoken modalities. (1.6) provides some examples: (1.6) derxt ha xub ha doxtr ha tree PL good PL gir l PL

PAGE 21

21 zxmi ha geda ha prnde ha injured PL poor PL bird PL shir ha medad ha divar ha lion PL pencil PL wall PL miz ha rud ha xane ha desk PL river PL house PL While all borrowed nouns are pluralized by the suffix / ha/, some borrowed Arabic nouns are used in their broken plural forms. These borrowed Arabic nouns are mostly used in formal and written language. (1.7) fkar fekr ha ha thought.PL thought PL poem.PL poem PL fks ha kopi ha tank ha bomb ha fax PL copy PL tank PL bomb PL These facts about pluralization in Persian are important because all of the materials in the experiments presented in this dissertation use inanimate plural nouns.

PAGE 22

22 Further, in Chapter 7 I will present the results of an experiment that explicitly tested the impact of these different morphological markers of plurality on agreement phenomena in Persian. Simple and Compound Verbs in Persian Turning now to verbs in Persian, Dabir Moghaddam (1997) classifies Persian verbs into two categorie s: simple and compound. Simple verbs consist of a verbal constituent while compound verbs consist of a non verbal constituent (a noun, adjective, past participle, prepositional phrase, or adverb) plus a verbal constituent. The verbal and non verbal consti tuents of the compound verbs are independent. Both compound and simple verbs are commonly used in modern Persian. But, in many cases, simple verbs are used more often in formal and written language (1.8) Compound verbs Simple verbs xonk krdn nushidn cold to do to drink gush krdn shenidn ear to do to hear neshan dadn pridn show to give to jump

PAGE 23

23 Unlike pluralization in nouns, whic h is largely agreed on, there is some controversy about how to characterize complex and simple verbs in Persian. Dabir Moghaddam does not agree with Mohammad & Karimi (1992: 201), who believe that the verbal element of complex verbs are best characterized as light verbs and as such are semantically empty, require the nominal element to provide semantic content and do not bear a thematic relation to the nominal element. In this research, I will use Dabir ecause his approach clearly vs. /kotk According to him, the verbal element expresses the mode of activity and the non verbal element expresses activity itse lf. Looking at t hese two compound verbs, we can see that the nonverbal parts are the same and the verbal parts are different. So, the difference of meaning and the thematic role the verbs assign to their subject is in the verbal part. As it is explained be low with examples (1.21 1.23), Dabir arguments account for subt le differences of interpretation in compound verbs, such as in the case of / nfs vs. /nfs keshidn verbal noun but alternative forms of ver bal part supports the Dabir

PAGE 24

24 Dabir Moghaddam classifies the Persian compound verbs as either: (1) Combination or (2) Incorporation. In the target items of the fiv e experiments presented here, I have used combined compound verbs. Using combined compound verbs (and not simple verbs) of Persian in the completion tasks (explained in the method section of each experiment) gave me the best opportunity to provide the part icipants with the preambles including the non verbal part of the compound verbs requesting them to complete the sentence with the verbal part which includes the tense, aktionart, and verb ending (which is the suffix marker of subject verb agreement). Comb ination In combination type verbs, the non verbal constituent (a noun, adjective, past participle, prepositional phrase, or adverb) combines with a verbal constituent. The meaning of the combined compound verbs (except for adjective + auxiliary combined c ompounds) is not transparent and in some cases is actually metaphoric. Combined compound verbs are not highly productive (except for adjective + auxiliary combined compounds) Below are examples of each type: Adjective + a uxiliary In this open set, an ad jective is combined with the stative transparent and it is very productive. (1.9) h va a ftabi bud weather sunny be.PAST.3SG

PAGE 25

25 (1.10) ab grm shod water warm become.PAST.3SG (1.11) (anha) kiyan o x brdar krd n (they) Kiyan OM informed make.PAST.3PL Noun + v erb This is not an open set. For noun plus verb type verbs, a noun is combined with different simple verbs (e.g. /k (1.12) se sa kar krd m three hour work do.PAST.1SG (1.13) Ba u d st dad m With 3.SG hand give.PAST.1SG (1.14) Drs ha ro yad gereft Lesson Pl OM learning take.PAST.3SG

PAGE 26

26 Prepositional phrase + verb In this type of verb prepositional phrase is combined with a verb. This, too, is a closed set. (1.15) be yad av rd im to remembrance bring PAST.1PL Adverb + v erb In this case an adverb is combined with a verb. It is a closed set. (1.16) xmir ha vr amd dough Pl up come.PAST.3SG Past participle + passive auxiliary : In this case a past participle of a transitive verb is combined with the auxiliary verb /shod This produces a passive verb, and is productive in Persian; it is an open set. (1.17) b ste ha ferestade shod n parcel Pl sent become.PAST 3PL Incorporation In incorporation, either the direct object or some prepositional phrase in corporates with the verb and creates a compound verb. Through the incorporation the direct object loses its grammatical ending(s). Also, the prepositional phrases which function as adverbs of location lose their preposition. Then the object or the noun of the prepositional phrase incorporate with the verb. (Dabir Moghaddam 1997: 26) Incorporation of d irect o bject through this process the direct object loses its grammatical endings (such as the postposition ra, the indefinite marker / i/, the plural suffi x, the possessive pronominal suffix) and becomes like a [ specific, definite object]

PAGE 27

27 and incorporates with the verb .The created verb is an intransitive compound which is Moghaddam 1997: 41). The meaning of incorporation compoun d verbs is transparent. (1.18) a. Niki zrf ha ro shost Niki dish Pl OM wash.PAST.3SG b. Niki zrf shost Niki dish wash.PAST.3SG Incorporation of prepositional phrase Some prepositional phrases function ing as adverbs of locations may incorporate with the verb, losing the preposition in the process. (1.19) a. ayene be z min oftad mirror to ground fall.PAST.3SG b. ayene z min oftad mirror ground fall.PAST.3SG Dabir Moghaddam (1997:46 47) explains the differences of combined and incorporated compound verbs. All the incorporated compounds have a non incorporated

PAGE 28

28 counterpart. They are intransitive (the incorporation reduces the valency of the verb). T hey are very productive, and lexically and semantically transparent. transitive or intransitive, as the combination process does not change the valency of the verb. They are neither as productive as incorporated compounds nor are they transparent (usually involving metaphoric extension). In the noun verb combined example, Dabir Moghaddam (1997: 46 47) argues that the existence of some combined compounds with the same non verbal noun but alternative forms of verbal part supports ktionart show the manner of the realization of the event in the way that the speakers perceive and conceptualize them. (1.20) a. nfs zdn breath to strike b. nfs keshidn Breath t o pull Dabir Moghad da

PAGE 29

29 Providi vs. /kot compoun Moghaddam believes that in these compound verbs the simple verb acts as an aktionart marker, as it has been lexicalized. Hence, the HAVE ing viewpoin Based on his explanation, it is easy to explain that the compound with the verbal ING viewpoint (the subject of the sentence receives the action, so it is almost always acted on and the thematic role is patient). Below are some examples (from Dabir Moghaddam 1997: 46): (1.21) gul xordn gul zdn deceive to eat deceive to strike (1.22) sili xordn sili zdn slap to eat slap to strike

PAGE 30

30 (1.23) ta xordn ta zdn fold to eat fold to strike Dabir Moghaddam (1997:47) explains that in line with Binnick (1991:148,170), he and Passive and Inchoative in Persian Passive and inchoative are structures that I used in the materials for Experiments 1 4. Persian grammarians and linguists have had different ideas about passive structure in Persian. This could be due to the similarities of these two structures: passive and research I have followed Dabir and Inchoative verbs. Passive A Persian passive sentence consists of a subject noun, past participle, and the ken place sentences.

PAGE 31

31 (1.24) cheraq ha (tvsote bche ha) shekste shod n light PL (by child PL) seen became.3PL Inchoative Th e inchoative structure consists of a subject noun, an adjective, and the itself (i.e. the inchoative reading) without the interference of an agent. (1.25) hva srd shod weather cold became.3SG With the following examples, Dabir Moghaddam (1982:79) explains about a group of sentences with inchoative structure (subject + adjective + the auxiliary and passive reading. The sentence (a) is an active sentence in which the subject is the agent of the verb. Sentence (b) is an inchoative sentence because the action happens by its own and there is no agent. Sentence (c) has inchoative structure but the phrase (by neighbors) indicates that the action has been taken deliberately. The sentences (b) two se ntences, the sentence (d) will be produced whose structure is similar to inchoative (adjective + the auxiliary), but the action could be deliberately or on its own accord. Dabir

PAGE 32

32 (1.26) a. hmsaye ha mashin ra pnchr krd n neighbor PL car OM flat made 3PL b. mashin (xod be xod) pnchr shod car (self with self) flat became.3SG c. mashin (tvsot e hmsaye ha) pnchr shod car (by neighbor PL ) flat became.3SG d. mashin pnchr shod car flat became.3SG Subjec t Verb Agreement in Persian The most important feature of Persian for our present purposes is the way in which Subject Verb agreement functions. According to Mahootian (1997), Persian is a Nominative Accusative language. The subjects of both transitive and intransitive verbs are unmarked for case. No morphological distinction exists between agentive and non agentive subjects. In this pro drop language, the subject is optional and the verb endings agree with the subject of the sentence. The verb endings used in formal and informal Persian are shown below. In formal Persian, when the stem ends in {/t/ or /d/},

PAGE 33

33 the verb ending for 3SG is [] zero morpheme. The verb ending / d/ is used elsewhere. (1.27) Person formal verb ending informal verb ending 1SG / m/ / m/ 2SG / i/ / i/ 3SG / d/ (present), (past) / e/ (present), (past) 1PL / im/ / im/ 2PL / id/ / in/ 3PL / nd/ / n/ There is person and number agreement between the subject and the verb; the verb conjugates for 1, 2, and 3 persons, singular (one person/entity) or plural (more than one person/entity). There is no agreement for gender as it is not morphologically encoded in Persian (neither for nouns, adjectives not even for pronouns). (1.28) (mn) kelid ra be u mi deh m (I) key OM to 3.SG IND give.PRES 1SG (1.29) (to) kelid ra be u mi deh i (you) key OM to 3.SG IND give.PRES 2SG

PAGE 34

34 (1.30) (u) kelid ra be Niki mi deh d (she/he) key OM to Niki IND give.PRES 3SG (1.31) (ma) kelid ra be u mi deh im (we) key OM to 3.SG IND give.PRES 1PL (1.32) (shoma) kelid ra be u mi deh id (you) key OM to 3.SG IND give.PRES 2PL (1.33) (anha) kelid ra be u mi deh nd (they) key OM to 3.SG IND give.PRES 3PL means that if the subject of the sentence is inanimate third person plural the subject verb agreement for number is optional and the verb can be either third person singular or third person plural and both forms are grammatical. This optionality of subje ct verb agreement is only for inanimate third person plural nouns and not for any other persons or animates. Below are some examples:

PAGE 35

35 (1.34) bche ha xis shod nd /*shod child PL wet became 3PL /became.3SG (1.35) brge ha xis shod nd /shod paper PL wet became 3PL / became.3SG (1.36) kargr an kar mi kon nd /*mi kon d worker PL work IND do 3PL /IND do 3SG (1.37) cheraq ha mi derxsh nd /mi derxsh d. lamp PL IND shine 3PL /IND shine 3SG (1.38) bazigr an hme ra mi trsan nd /*mi trsan d Actor PL all OM IND frighten 3PL /IND frighten 3SG (1.39) fekr ha ysh hme ra mi trsan nd /mi trsan d thought PL his all OM IND frighten 3PL /IND frighten 3SG

PAGE 36

36 the above description: which agree with the subject in person and number. The subject must be coded on the verb even if the subject NP is pro dropped. An important exception to subject verb agreement is with inanimate plural subjects, which can take a Providing t he following example, Mahootian (1997:136) explains that the choice of plural or singular verb with the inanimate plural subject is optional: (1.40) chmedun ha tu ye mashin e suitcase PL in EZ car is (1.41) chmedun ha tu ye mashin nd suitcase PL in EZ car are Proba ble Factors Influencing Agreement in Persian As one might imagine, there has been some interest within Persian linguists regarding the features that influence or determine optional agreement. For example, Sedighi (2005:3) argues that animacy in Persian has morphological exponents and forces a restriction on subject verb agreement. In the following examples, she shows that animacy is morphologically realized because the se lection of plural marker depends on the animacy of the subject. The plural suffix / an/ is used only for animate plural nouns while the suffix / ha/ is used both for animate and inanimate plural nouns.

PAGE 37

37 (1.42) mn doxtr an ra did m I girl PL OM see.PAST 1SG (1.43) mn ketab ha ra did m I book PL OM see.PAST 1SG Sedighi (2005) writes that Thacktson (1978), Meshkat al dini (1987), and Saadat (1996) have pointed out the optionality of the agreement but no syntactic o r morphological explanation has been provided in this regard. Saadat (1996) assumes that the reason for the constraint is initiated from classical Persian in which the inanimate subjects were not considered as the real agents as they did not have control o ver their actions. Saad at (1996) provides a series of examples from literature texts showing that the verbs of the inanimate nous in classic texts were only singular. He believes that the reason for the mismatched form of the verb is that the inanimates we re considered as a unit and not individuals. But for a f ew exceptional examples from literature texts in which the inanimate plural nouns have plural verbs, he explains that the inanimate nouns can have plural verbs in case they are used metaphorically and personified. Meshkat al dini (1987) believes that the optional agreement depended on the Individuality vs. Unity of the inanimate plural subject. If the emphasis is on the Individuality of the entities of the subject, the verb is plural but if the empha sis is on the Unity of the entities of the subject, the verb is singular. Similarly, Windfuhr (1979) states

PAGE 38

38 that implicitly or explicitly human subjects will show agreement with a verb, while, the subject verb agreement for non human subjects indicates the emphasis on the number of individual items. Sedighi (2005) proposed an account using a Distributed Morphology framework to explain the optional inanimate plural subject verb agreement in Persian. Based on Distributed Morphology, syntax is the place wher e the subject verb agreement occurs. Sedighi (2005, 2007) argues that it is after syntax and before vocabulary insertion that the impoverishment rule deletes the number feature and causes the emergence of a less marked feature which is singular form. In th is account, inanimate subjects create a constraint on the subject verb agreement so that the verbs can be less marked (singular) or plural. She proposes an optional impoverishment rule by which the number feature is deleted when the subject has [ animate] feature. Sedighi asserts that the factors of this optional impoverishment or situations in which it is applied are not clear yet. In yet another account, Lotfi (2006) argues that the use of singular verb for inanimate plural subject indicates a collective conceptualization of the plural inanimate subject i.e. the entities of the subject are considered as a single unit. While the use of plural verb indicates a distributive conceptualization, i.e. entities of the subject are considered individuated. Collecti ve and distributive conceptualizations do not mean that they are necessarily collective and distributive nouns. The Persian collective nouns such as / glle/ herd tim/ team inherently collective in the lexicon. Lotfi explains that the collective and distributive conceptualizations are morphologically plural and context sensitive; in some contexts

PAGE 39

39 they have plural verbs while in others the verbs are singular. Distinguishing between singular verbs if they are collectivized in action but they get plural verb if they are individuated in the action. With the following example he argues that autonomy brings about the ch oice between singular and plural verb endings. The singular verb is used when the entities of (1.44) sosis a suxt (n lubia ha) sausage PL burt.3SG (not bean PL (1.45) sosis a suxt n sausage PL burnt 3PL In conclusion, Lo tfi (2006:136 137) states that: the verb is the unmarked linguistic rea lization of an action or event [ so ] the proximity between the verb and the conceived autonomy of members of a group in action makes verbs natural markers of col lectivity and distributi veness. He suggests further research to investigate the effect of verb type, tense, and aspect believes that Persian speakers prefer to use plural verbs when the tense of the

PAGE 40

40 sentence is future and not past. It may seem more acceptable to have autonomy in actions when the course of events has not finished yet. followed by any systematic experiment investigating the effect of the factors or by further explanation/ example in this regard (2007) conducted a study with three tasks. In the preference task, the participants were instructed to read a set of twenty se ntences with different structure (which included inan imate plural subjects) and indicate whether the singular or plural form of the verb or both forms were appropriate. The results showed that the participants largely preferred one of the forms in the expe riment either singular or plural and did not accept both forms. Based on this result, they suggested that singular and plural forms of the verb are not in free variation and there should be some significant factors for the production of SG/plural forms of the verbs. In another task, ten scenarios with different structures were provided and the participants were requested to complete the scenarios. The results showed that different proportions of singular and plural verbs were used by the participants. The authors suggest that the use of singular verb could be due to the lack of individuation showed that more plural verbs were used when the subjects showed animate featur es such as movement. Also, the participants used more plural verbs for the subjects such as houses, shops and balloons that were portrayed very distinct from each other. Sharifian & Lotfi (2007) suggest that the choice of singular vs. plural is at least t o zation. If the speaker conceptualizes

PAGE 41

41 the subject of the sentence as a whole (low level of construal resolution), the verb is ect, the verb will be plural. They conclude that the optional use of singular vs. plural to express the same fact is in fact using different forms capturing different construal of the experience. However, it is important to note that Sharifian & Lotfi (200 7) did not specifically test the influence of any oth er factors in the subject verb optional number agreement. Further, the structures of the target sentences were not controlled for and it remains unclear what other factors may be relevant for Persian. Th e Goal and O utline of the Dissertation As Sedighi (2007) has asserted and as far as I have found ,no systematic study has been accomplished to test the effects of any specific factor influencing agreement patterns in Persian. Hence, the situations in which the Persian speakers/writers prefer to use singular vs. plural verbs for the inanimate, plural nouns are unknown, despite a great deal of speculation and argumentation. Thus, further investigation is needed to test the effect of different factors influenc ing the optional subject verb number agreement. My research goal was to determine the effect of grammatical and conceptual factors in the production of number agreement in Persian, and apply these results to better understand both how such agreement is re presented and produced. Based on prior theoretical and experimental research in Persian and other languages, I identified the following factors as the most potentially relevant, and tested them across five experiments, presented in chapters three to seven. As a native speaker of Persian, I created and prepared all the target and distracter items of the five experiments of this dissertation (listed and glossed in Appendices A to E at the end of the dissertation).

PAGE 42

42 Chapter 3: In Experiment 1, a judgment task w as used to investigate (a) the acceptability of the sentences with singular and plural verbs testing the effect of verb tense (e.g. past, present) and the thematic role of the subject nouns, (b) the role interpretation of the inanimate plural subject nouns in the same target sentences testing the effect of verb tense (e.g. past, present), the thematic role of the subject nouns, and verb number (singular, plural). The materials of this experiment are the complete sentences of the preambles used in experiment 2 for completion. Chapter 4: In Experiment 2, a completion task was used to test the effect of verb tense and the thematic role of the subject nouns(e.g. agent, patient). The goal of this experiment was to test if speakers would produce more singular verb forms with plural subjects when the verb tense was past or present and the thematic role of the subject was agent, patient, theme or instrument. Chapter 5: Experiment 3 tested the effect of conceptualization of the entities of the subject noun as individ ualized vs. unified. More specifically, this experime nt was to test whether the speakers would produce more mismatched verb forms with individualized vs. unified plural subjects. Chapter 6: Experiment 4 tested the effect of concreteness vs. abstractness of the subject noun s, i.e. to test whether the speakers would produce more mismatched verb forms with concrete vs. abstract subject nouns. Chapter 7: Experiment 5 tested the attraction effect of object nouns in the subject object verb word order (SOV) se ntences In this experiment the effect of object local noun, its animacy, singularity or plurality, and its different plural morphemes (the

PAGE 43

43 unspecific plural morpheme / ha:/ and the specific animate plural morpheme / an/) in the produced form of the verb were tested. In all experiments except Experiment 1, participants performed a verbal sentence completion task in which they read the beginnings of sentences and then completed the sentence aloud. I coded and analyzed the proportion of mismatched (singular ) verbs in their responses to test the effect of the mentioned factors in the optional agreement. The Importance of the Study The focus of this dissertation is on the optional morphological agreement between subject nouns and verbs in Persian language. Whi le number agreement may seem to be somewhat esoteric, it has been of great interest to linguists and psycholinguists because agreement phenomena are quite common in language and can provide an important evidence about how language is structured and how tha t structure is treated during language production and processing. As we shall see in Chapter 2, there has been a great deal of research on subject verb agreement in other languages, all of which require subject verb agreement. To date, there has been no e ffect to either systematically test the claims of Persian linguists, or to apply insights from Persian to psycholinguistics theories of subject verb agreement. Over the past 15 years, two major approaches to agreement processing have emerged: The Minimali st approach, in which agreement is governed solely by syntactic information, and The Maximalist approach, in which both conceptual and syntactic information influence agreement during grammatical encoding. There is empirical evidence for both approaches, h owever, the data have always come from languages in which verb agreement is required, and so theories have necessarily relied on data from

PAGE 44

44 structures that are ungrammatical. However, because it is possible to have grammatical ian, this allowed me to study agreement without requiring speakers to respond to or produce ungrammatical sentences. In fact, Persian has a number of characteristics that make it an informative case: First, subject verb agreement studies in other languages have necessarily required using elicitation of errors and attraction effect techniques to test the effect of syntactic and non syntactic factors in agreement. Thus, the material of the experiments were sometimes ambiguous sentences and the collected data were ungrammatically produced sentences with agreement errors. The optionality of subject verb number agreement in Persian offers an excellent opportunity to study subject verb number agreement using grammatical Persian sentences. There is no need to use e licitation of agreement errors in the experiments and use ungrammatical sentences and agreement errors as data because both matched (plural form) and mismatched (singular form) of the verbs are grammatical and acceptable. Second, previous studies in agreem ent have used a completion task in which the participants were provided with a series of preambles and they were supposed to repeat the preamble and complete the sentence with their own words. The participants were free to complete the sentence with any ve rb/structure. But, using different verbs/structures could be an interfering factor in the experiments affecting agreement errors. Using compound verbs of Persian gives us a golden opportunity to avoid this factor. The non verbal part of the compound verbs (as a part of the preambles) were given to the participants and they were requested to complete the sentence with their own verbs. In this way the participants (not knowingly) were restricted to use the verb

PAGE 45

45 that was designed to complete a preamble with. S o, I had this opportunity to test the effect of verb type in the agreement and keep the effect of verb type constant in all the experiments. Third, previous studies have tested the effect of morphology in the agreement using irregular/invariant plural nou ns, and sometimes because of the restrictions of languages they could not test this factor alone (they had to mix this factor with another factor, e.g., gender). Again, Persian has a property which can be very helpful to test the effect of morphology. Havi ng two plural morphemes / ha/ (for all nouns) and / an/ (specifically for animate nouns) provided me with a very exceptional opportunity to test the effect of morphology in agreement. Finally, previous studies have tested the attraction effect of a local noun interfering between subject head noun and verb. In most of the experiments, this local noun was the noun phrase of the prepositional phrase modifying the subject head noun (so the local noun was a part of the subject phrase). In some experiments, the local noun was the noun phrase of a relative clause. In a few experiments in languages like French, the local noun was an object clitic pronoun (not noun) located between the subject noun and the verb. In fact, the word order of the studied languages were SVO and thus researchers could not test the intervening attraction effect of object noun as a local noun. Persian, like any SOV language, can be useful because the word order of unmarked sentences is SOV, and so the object falls naturally between the subj ect and verb. This provides a way to test the attraction effect of main object nouns in the optional subject verb agreement.

PAGE 46

46 The combined pattern of results in this dissertation supports a constraint based, Maximalist approach to number agreement, particul arly by demonstrating effects of thematic role, concreteness vs. abstractness, and unity vs. individuality of the subject nouns in number agreement. In my dissertation, I will propose a model for agreement that draws on prior theoretical and experimental d ata from other languages in addition to my da ta from Persian. Because no other studies of sentence production in Persian have been published, the work presented in my dissertation provides a set of unique and important data for researchers interested in a language with grammatical number mismatches and allows me to create a model that is more wide reaching in scope.

PAGE 47

47 CHAPTER 2 R EVIEW OF RELATED LITERATURE Overview As discussed in Chapter 1, in many languages, the main verb in a sentence must be inflect examples of this in English are given in (2.1) below, in which the form of the verb walk changes (with the addition of s for the singular subject Mary un dergo more dramatic suppletion processes, as seen in (2.1c) and (2.1d). (2.1) a. Mary walks every day b. Mary and John walk every day. c. Mary is a student. d. Mary and John are students. A considerable amount of research has explored the mechanism of subject verb number agreement during language processing, production and comprehension. This increasing interest in the processing of agreement is due in part to the prevalent nature of agreement cross linguistically. According to Mallinson & Blake (198 1), three fourths of languages have agreement which is overtly marked in most, if not all, of the sentences. Because agreement is a fundamental aspect of languages, understanding the mechanism of agreement is an important part of understanding speech produ ction. This chapter includes a summary of the studies in the production of subject verb agreement, starting with different theories about subject verb agreement in sentence production and evidence for them, and then going on to discuss the results more gen erally from experiments that have tested various factors on agreement.

PAGE 48

48 Psycholinguistic Accounts of Subject Verb Agreement Processing Accuracy and efficiency are two essential properties of language production that theories of language production should explain. On average, speakers of different languages produce 2 3 words per second. Speakers retrieve these words from their lexicon, which has about 30,000 words. In the process of speech, speakers must not only retrieve words, but also construct syntactic relations among these words (such as number agreement between the subject and verb or gender agreement between a noun and adjective). The speaker needs to combine the information retrieved from memory to the phonological forms (words) with correct pronunc iation. Although this task is very complex, the speakers use language very fast and easily. According to Bock (1991), slips of tongue occur approximately once every 1,000 words, as estimated from the London Lund corpus (Garnham, Shillcock, Brown, Mill, & C utler, 1981). Vigliocco & Frank (1999 & 2001) classify theories of agreement into two classes based on the issues of accuracy and efficiency. These two classes of theories of agreement can be thought of (and are labeled by Vigliocco & Frank) as maximalist and minimalist These approaches are considered mutually exclusive, as the assumptions of one class contradict the assumption of the other. According to Vigliocco & Hartsuiker (2002: 443), the minimalist approach follows the principles of modularity, and most importantly, the assumption of information encapsulation in which independently modules do not have access to facts that other systems know about (Fodor, 1983). The two main assumptions of the minimalist approach are 1) minimal input and 2) uni direct ionality Minimal input means that only the minimum necessary information flows among the levels and thus any non necessary information from an earlier level cannot influence processing at the next level. It was for this reason that Vigliocco & Frank

PAGE 49

49 (1999 directionality means that information from a later level cannot influence processing at a prior level via feedback the minimalist approach thus is a feed forward model. In the minimalist approach, accuracy is att ained by insulating each process at one level. This means that syntactic encoding is insulated from non necessary information at conceptual and phonological levels that could interfere. Efficiency is also attained by the properties of modular systems. Sinc e modules are narrowly focused, they can be fast in their functioning. According to Vigliocco & Hartsuiker (2002: 443), quoting from (Boland & Cutler, 1996), the other class of theories (maximalist) These theor ies assume 1) interactivity of information : different types of information (syntactic and non syntactic) interact in the process of agreement and 2) maximal input : the common information from other levels can be used in other levels, even the levels after them. In fact, it was for this reason that Vigliocco & Frank (1999) label this This approach achieves accuracy by using the converging information available to the system. In this way, related information from different layers ca n be used to compensate loss of information. This related information could be of the same type (syntactic) or of different type (semantic). This view achieves efficiency through maximal input This means that the availability of maximal input before word selection gives the opportunity for faster encoding by virtue of preactivation (e.g. cascading of activation) Another factor which assists efficiency is bidirectionality of information, that is, information can be used by either prior or later processes.

PAGE 50

50 Evidence for Minimalism In a series of completion tasks, Bock and Miller (1991) used elicitation of errors to examine the process of subject verb agreement. In these experiments, the participants were given a series of subject preambles and they were as ked to repeat and complete the sentence. The preamble consisted of two noun phrases (e.g., the key(s) to the cabinet(s) ... ). One of the nouns was a subject head noun (e.g. key or keys) and the other noun was the noun of the prepositional phrase (e.g. cab inet or cabinets head noun and the verb and had an interfering role in the subject verb agreement. They found that more agreement errors were produced when the head noun and local noun mismatched for number. There were significantly more errors when the head noun was singular and the local noun was plural. Bock & Miller (1991) also tested the effect of notional number. They prepared a set of target preambles in which the head nouns singular but referring to multiple tokens, e.g., or single token nouns (grammatically singular and referring to one token, e.g., the letter form the lawyer The result showed no effect of distributivity with equal agreement errors found for both the distributive and single token subject head nouns. The effect of semantic factors like animacy in subject verb agreement was also tested by Bock & Miller (1991), and by Bock, Eberhard, and Cutt ing (1992). Their results showed that animacy played a role in selecting the noun phrase as a subject but did not affect error rates in subject verb agreement. The authors thus concluded that animacy does not have any significant effect on subject verb num ber agreement. Bock & Eberhard (1993) examined the effect of phonological factors on the production of

PAGE 51

51 subject verb agreement in English. Their results showed no effect of phonological factors on agreement. Bock & Miller (1991) and Bock & Eberhard (1993) c oncluded that syntactic feature processing is the mechanism underlying the agreement errors and the process is during a strictly grammatical stage of the production process: Agreement is only affected by the syntactic properties of the subject head and the local nouns. Semantic interpretation of the subject and morpho phonological information do not have any effect on the subject verb agreement. According to Vigliocco, Butterworth, & Semeza (1995: 188), the minimalist view is consistent with many linguisti c treatments of agreement ( Chomsky 1965; Akmajian & Heny, 1975; Gazdar et al., 1985). They believe that in subject verb agreement the features of a subject (person, number, gender), as the source or controller, is copied to the verb, which is the target. This view is also consistent with the psycholinguistic models of speech production that accept copying operations (Garrett, 1980; Kempen & Hoenkamp, 1987; Levelt, 1989). Vigliocco, Butterworth, & Semenza (1995: 188) quoted some speech examples from (Stemb erger, 1985: 154) that support this view that agreement is computed after lexical heads have been exchanged. In these examples, subject verb agreement is (2.2) a. Most c ities are true of that. (intended meaning: that is true of most cities) b. The minimalist view is consistent with the hierarchical encapsulated model in which production processes are se parated into several distinct processing stages. The model I will discuss next is the one presented in Bock and Levelt (1994), one of the best

PAGE 52

52 known of the hierarchical encapsulated models. This model consists of three stages or levels. The first stage ( me ssage ) is the formulation of a non linguistic conceptual representation. The second stage ( grammatical encoding ) includes both the selection of syntactic framework. Du ring this stage, each lexical concept is mapped onto its abstract syntactic representation and the syntactic part of speech planning (including the computation of agreement) takes place. Phonological encoding is the third stage of the model during which th e actual sounds to be produced are spelled out. Bock and Levelt (1994: 946 948) explain that the Message captures features of grammatical encoding. Grammatical encoding is g rouped into two sets, functional and positional processing Lexical selection is the primary subcomponents of functional processing which involves the identification of lexical concepts and the lemmas that are suitable for Lemmas carry the grammatical information associated with individual lexical concepts, such as their form class (noun, verb, etc.). Function assignment involves the assignment of grammatical roles or syntactic function (e.g., subject nominative, object dat ive). Positional processing fixes the order of the elements in an utterance. Constituent assembly is the creation of a control hierarchy for phrasal constituents that manages the order of word production and captures dependencies among syntactic functions. Inflection involves the generation of fine grained details at the lowest levels of this structure. Finally, phonological encoding is the stage in which the phonological content of the utterance is spelled out.

PAGE 53

53 According to Bock and Levelt (1994), in this model, the flow of information between the levels is unidirectional, i.e. the information flows from message to phonological encoding. The system is modular ; it means that there are specialized processing mechanisms to carry out particular processes, and t hese processes are unaffected by information elsewhere in the system (dissociation among the use of semantic, syntactic, and phonological information). Based on this model, agreement is a syntactic process and as such is not affected by semantic factors. Semantic and conceptual information affect message formulation and the mapping from the message to the grammatical stage of production. Similarly, the morphological or phonological form of the lexical head will not feedback to influence agreement with the verb (the phonological encoding is after the grammatical encoding and cannot affect agreement process). So, the results reported by Bock and colleagues are consistent with the encapsulated hierarchical model. According to Vigliocco et al. (1995: 188), in this model agreement works in the following way: First, the number feature of the noun is formed, then the feature is just copied to the verb. Then the correct morpho phonological form of the noun and the verb are retrieved in the following stage. The pho nological form of the noun is not able to affect the production of agreement because the Number (of verb) has been computed in the previous stage. ( Figure 2 1 ) Evidence for Maximalism Many recent studies on agreement have showed clear effects of non syntac tic factors and have thus challenged the idea of the encapsulation of syntax in subject verb agreement production. The effects of semantic factors on subject verb agreement have been reported by many researchers. Among many studies, we can refer to the stu dies

PAGE 54

54 that Vigliocco and colleagues (Vigliocco, Butterworth, & Semenza, 1995; Vigliocco, Butterworth, & Garrett, 1996a; Vigliocco, Hartsuiker, Jarema, & Kolk, 1996b) conducted in Italian, Spanish, Dutch, and French that demonstrated the effect of distributi vity of subject head nouns on agreement. Eberhard (1999) demonstrated the effects of distributivity in English sentences when the subject head noun and the local nouns are imageable/concrete nouns. Bock et al. (1999) found effects of collectivity on subje ct pronoun agreement. Vigliocco and Franck (1999, 2001) reported the effect of conceptual information on gender agreement in Italian and French. Vigliocco et al. (1995) showed an effect of morpho phonological influences on agreement processing in Italian ( in which more errors occurred with invariant nouns that gender agreement errors were more common with nouns carrying ambiguous gender marking. And finally, Thornton and MacDonald (2003) showed effects of plausibility on agreement production in English. The above mentioned evidence showed that both syntactic and non syntactic information are influential in agreement production. It could be that conceptual information plays a more effective role in forming agreement. Hence, Vigliocco and colleagues (Vigliocco & Franck, 1999; Vigliocco & Franck, 2001; Vigliocco & Hartsuiker, 2002; Vigliocco & Zilli, 1999) argue that the production system utilizes all available informatio n to aid in the production of agreement. Vigliocco & Frank (1999) labeled this general view as maximalist In this framework, accuracy is achieved by using additional information available to the system

PAGE 55

55 which could be useful because it protects against in formation loss. Unnecessary conceptual information can strengthen the syntactic information. Vigliocco and Frank (1999) mention that this approach is compatible with computational psycholinguistic models that use a unification mechanism for syntactic struc ture building. According to Vigliocco, Butterworth, & Garrett (1996 a: 266 267), agreement is a relationship for number, person, and/or gender between two (or more) elements in a sentence. In subject verb number agreement, the number person gender relatio nship is between the subject and its verb. Based on the Minimalist approach, this But, computational psycholinguistic models According to this approach, each of the two elements involved in the agreement relation express some information about that relation. Through unification, the information of these two elements (e.g. subject and verb) is merged and creates the agreement rel ation. So, based on this view, features are unified (i.e. they are not copied or transported from one element to another) and as a result they can be shared by elements of different branches of syntactic structure. This implies that the elements (e.g. sub jects or verbs) have information about the structures they can potentially be combined with, in addition to the information about their meaning and phonology. Unification is not directional in nature because the information of the involved elements is merg ed. This point is particularly useful when thinking about Persian, which I will turn to after the next section.

PAGE 56

56 Current Psycholinguistic Accounts and Theor ies Linguistic theories have different views concerning agreement. Chomsky (1965) considered agreemen t to be the rewriting of terminal symbols. In this view, agreement rules were treated as similar to rules of assimilation in phonology. In this way, agreement has a minor role in the syntax. But, other accounts Gazdar et al. (1985) and Pollard and Sag (199 4) treated agreement as the core of syntax. There are different positions in psycholinguistics that explain about the mechanism of agreement in language production. Bock and Middleton (2011) mention five psycholinguistic accounts; control account with Mar king and Morphing model (Eberhard et al. 2005), constraint account by Pollard and Sag (1994), semantic integration hypothesis proposed by Solomon and Pearlmutter (2004), production syntax (Franck et al. 2006), and memory retrieval account proposed by Badec ker and Kuminiak (2007) Here, I compare the first two accounts (the control account and constraint account) which are related to the psycholinguistic approaches to agreement discussed earlier (Maximalist and Minimalist). In order to relate formal system s of linguistics to psycholinguistic issues, Bock & transmission of features within syn tactic structures. According to these accounts, subject number controls verb number. The number feature of the subject (the agreement controller) is transmitted to the verb (the agreement target) and there is no other syntactic mediation between them. Cons

PAGE 57

57 is based on semantic structures, is used by both subject and verb. So, the verb number (similar to subject number) is formed by meaning and consequently the verb number is not copied from the subject to the verb. Control accounts assume that the subject receives the number from the message. Then the subject plays a mediation role and transports the number feature to the verb. Because the verb does not have a direct connection with the message, it cannot get the number feature directly. So, first the subject creates its numb er, then the grammatical number of the subject (singular or plural) is transported to the verb. On the other hand, the constraint view assumes that both subjects and verbs have a direct connection to the notional number in the message and that they create (inflect) their number based on the notional number in the message. So, the retrieval and inflection of subject number and verb number happen independently. The notional co uld be the same or different. Bock & Middleton (2011) argue that if the subject number differs from verb number, then a unification process will align the differing features. ( Figure 2 2 ) Linguistic and psycholinguistic approaches to agreement have represe nted both the control and constraint accounts. In linguistics, generative grammar from its first ct features through control mechanism. There are also formal constraint mechanisms for agreement. Bock & Middleton (2011) discuss Pollard and Sag (1994) in this regard.

PAGE 58

58 According to Pollard and Sag (1994), the subject and verb carry independent values that are constrained by referential number on their indices. Agreement is the outcome of a greement. The Marking and Morphing model (Eberhard et al. 2005) is represented in a control theory. According to Marking and Morphing, only subject noun phrases and pronoun phrases (crucially, not verbs) can get notional number values. Verb number reflect s the grammatical number of its controller (i.e. the subject noun phrase). Pronoun number reflects the notional number of its antecedent. Verb agreement shows a small and systematic effect of the notional number (e.g. distributivity, collectivity) of the s ubject (controller). Any attraction effect on number agreement is due to the grammatical number (not the notional number) of the attractor. This means that the notional number of the attractor does not have any effect in the attraction effect (similar to m inimalist view discussed earlie r). Figure 2 3 shows the model. ( Figure 2 3 ) On the other side, the constraint position has been tested by Haskell and MacDonald (2003) and Thornton and MacDonald (2003). Vigliocco and Hartsuiker (2002, 2005; see also Viglio cco and Franck 1999, 2001) endorse a related view under the heading of Maximalism. According to Maximalist view, meaning is scattered in the agreement process, and all the elements of the agreement have notional number features. Bock & Middleton (2011) ar gue that constraint based, maximalist approaches (not control approaches) can better explain notional agreement. Constraint based

PAGE 59

59 approaches emphasize the importance of meaning and notional number as a factor which interacts with the grammar in agreement. Because although the effect of notional number may be much less than grammatical number, it has been reported by many experiments. In fact, agreement is the outcome of both semantic and syntactic interaction. So, following Pollard and Sag (1994), constrai nt based approaches assume that the agreeing elements (e.g., subject and verb) involve a relationship in which both subjects and verbs will get information about reference. In fact, both subjects and verbs independently have access to notional number. In t his way the constraint view can explain the effects of notional number on agreement. Another advantage of the Constraint based view which Bock and Middleton (2011) discuss is that it can explain the attraction effect in the same framework as it does for n ormal agreement. The Constraint view holds that canonical agreement is the situation in which the typical semantic and grammatical constraints align for agreement (based on the meaning, structural distances, and morphological cues). But, variations such as attraction can happen if the semantic, structural, or morphological constraints are outside the normal ranges. This approach does not yet have a detailed theory or model of agreement production, but Thornton and MacDonald (2003) discuss how agreement may be computing from a constraint based perspective. According to this account, the process of subject verb agreement involves the integration of multiple sources of information as part of producing an inflected verb form. Different factors have different d egrees of influence in this process, based on the reliability of the constraints of a particular verb

PAGE 60

60 form. For example, the grammatical number of the noun is a stronger cue to verb agreement than conceptual number and so it has stronger influence. In fact according their account, the effect of grammatical number of the noun is so high that it may be considered as the dominant constraint that can hide the effect of less important factors. Thornton and MacDonald (2003) examined the effect of plausibility i n particular on agreement. But, they also discuss the role that other probabilistic constraints may play in agreement, including distributional information in the input, that is, the probabilistic patterning of linguistic information, including phonemes, s yllables, words, and phrases. In a constraint based framework, multiple cues are integrated in the production of subject verb agreement. By testing the effect of different factors on the optional subject verb agreement in Persian, I believe that the effect of some of the cues such as grammatical number is very important and may be universal in languages while the effect of some other cues may be weaker and sensitive to specific languages. Psycholinguistic A pproaches and Agreement in Persian Explaining the process of optional subject verb agreement in Persian is not easy. As explained in Chapter 1, in Persian the verb agrees with its subject for number and person. The verb endings are markers of person and number of the subject/agent of the sentence. Since P ersian is a pro drop language and the subject of the sentence can be dropped in most of the sentences, the role of verb endings for marking the agent/subject of the sentence is very important. Persian grammarians and linguists have reported the optionality of subject verb number agreement when the subject of the sentence is inanimate plural noun. This means that when the subject of the sentence is an inanimate plural noun, the related verb can be singular or plural and both forms of the verb are grammatical acceptable and widely used in spoken and written language.

PAGE 61

61 Minimalist and control account, as explained above emphasize the effect of syntactic factors in the process of agreement. Subject verb agreement is formed by copying the features of subject (as s ource) to verb (as target). Clearly, this approach and the related models of speech production may not be able to explain the optional subject verb number agreement of Persian. Because the Number feature of the subject nouns of these grammatical acceptable sentences is [plural] while the verbs could be [plural] or [singular]. It is worth noting again that all the experiments about the production of agreement have used attraction effects and elicitation of errors (ambiguous preambles/ ungrammatical data) in order to study the effect of syntactic and non syntactic factors in agreement. However, the mismatched form of the verb (optional agreement) in Persian are in grammatical and acceptable sentences. Even sentences without attraction effects can have optional agreement (singular or plural verb). I believe that maximalist approaches and a constraint based account are better suited to account for subject verb agreement in Persian. A constraint based, maximalist approach assumes the interaction of syntactic and non syntactic information during the computation of agreement. Hence, in addition to grammatical information, conceptual information is also taken into account. Also, it can explain the attraction effect of semantic, structural, or morphological constraint s outside of the normal range of the agreement. In this dissertation, through five experiments, I tested the effect of syntactic and non syntactic factors in the production of subject verb number agreement in Persian. Based on the provided evidence of the factors affecting subject verb agreement process, I believe the effect of grammatical and conceptual information along with the

PAGE 62

62 syntactic and morphological attraction effect of object nouns (as a non subject local noun) can be factors in the optional subje ct verb agreement found in Persian. Based on the constraint approach, conceptual information is predicted to be influential in computing agreement. So, in addition to the syntactic information from the subject, object and verb, conceptual information can a lso be considered as a crucial non syntactic factor involved in the optional subject verb agreement in Persian. This would be in line with the idea of unification used in the maximalist approach. That is, contrary to the minimalist approach, my hypothesis is that verb does not copy the number features from the subject and instead the number on the verb is the result of the unification of available information in different elements of the sentence. Syntactic Factors I nfluencing Subject Verb Production and Processing In this section I now turn to the factors that have been shown to have an impact on subject verb agreement, including reviewing prior experimental work discussed above in greater detail. Crucially, in all of the languages tested, subject verb ag reement is obligatory. Attraction Effects Research in psycholinguistics investigating subject verb agreement in production started with seminal work by Bock & Miller (1991), who conducted a series of experiments that elicited agreement errors under control led circumstances in order to examine the process of subject verb agreement. In these experiments, participants were provided with series of preambles that consisted of a subject head noun and another noun in a post modifying prepositional phrase. This int erfering noun, called the the head noun and local noun were systematically varied in order to explore the effect of

PAGE 63

63 the local noun on the production of subject verb number agreement. An example is shown below in (2.3) (2.3) a. Head SG Local SG (match): the key to the cabinet b. Head SG Local PL (mismatch): the key to the cabinets c. Head PL Local SG (mismatch): the keys to the cabinet d. Head PL Local PL (match): t he keys to the cabinets Participants were instructed to repeat the preamble and complete the sentences with their own words. The number of agreement errors that participants made was analyzed. The results from Bock and Miller showed that the rate of errors in the mismatched conditions was markedly higher than in the matched conditions. This pattern indicated that the local noun did indeed affect subject verb agreement. This verb to appear to agree with it instead of the subject noun. Interestingly, the attraction effect was larger when the head noun was singular and local noun was plural compared to the reversed mismatch in which the head noun was plural and local noun was singular. This attraction effect has been replicated and extended by many researchers since, including: Bock & Cutting (1992), Bock & Eberhard (1993), Eberhard (1997), and Eberhard (1999) in English; Vigliocco, Butterworth, & Semenza (1995) in Italia n; Vigliocco, Butterworth, & Garrett (1996a) in Spanish and English; Vigliocco, Hartsuiker, Jarema, & Kolk (1996b) in Dutch and French. All of these studies have reported an effect of attraction for local nouns, and much work has focused on determining the factors that increase agreement errors in speakers. Researchers have relied on the

PAGE 64

64 attraction effect of local nouns and used elicitation of agreement errors under almost equal conditions to test the effects of syntactic and non syntactic features on subje ct verb agreement. Only Syntactic Factors As we saw above, agreement is traditionally considered a syntactic phenomenon (e.g. Corbett, 1994). As such, it could be a solely syntactic process insulated from conceptual or other non syntactic influences (cf the discussion of minimalism above), and there is some experimental evidence supporting this view (e.g. Bock & Miller (1991), Bock & Eberhard (1993), and Vigliocco, Butterworth, & Garrett (1996a; Experiment 3 & 4). Continuing our discussion of Bock & Mil ler (1991), they also tested the effect of number, animacy, distributivity and length of the local, subject modifying phrase. In three completion task experiments, participants were provided with a series of the key(s) to the cabinet(s) and .The Number, Animacy, and Concreteness of the head noun and the noun at the post modifying prepositional phrase (i.e. the local noun) were manipulated and the number of head noun and local nouns were either matched or mismatched. While, as discussed above, there was an attraction effect, the animacy and concreteness of the local noun (e.g. the author(s) of the speech(es ) vs the speech(es) of the author (s )) did not show any effect on the rate of agreement errors. Then, they tested the role of animacy in sentences that also manipulated the the key(s) to the th and vs tamperproof The result showed no

PAGE 65

65 significant effect of the length on agreement errors. From this, Bock and Miller concluded that only syntactic features of subject (e.g. number) are influential in agreement, whil e notional properties (e.g. concreteness, animacy) and the positions of noun phrases do not have any effect. Vigliocco, Butterworth, & Garrett (1996a) (in Experiments 3 & 4) tested the attraction effect of distributive nouns in English, using a similar des ign and set of target preambles. These preambles consisted of a singular head noun and a plural local noun without a number matching control. There were two conditions: a single token condition (e.g., the letter form the lawyers ) and a multiple token condition (e.g., the label on the bottles ). In their Experiment 3, participants heard a series of these sentence preambles and repeated them and then completed each with th eir own words. The results showed that English speakers were not sensitive to the distributivity manipulation. Experiment 4 used the same materials as Experiment 3, but matched each of the preambles with a semantically plausible adjective. In each trial, the participants saw on the computer screen an adjective immediately followed by a sentence preamble (e.g. and then ). They were supposed to make up a question using the sentence beginning and the adjective. (e.g. ). The results of this experiment, similar to Experiment 3, did not show any distributivity effect and was constant with Bock and Miller (1991). The null effect of distributive number was also replicated b y Bock, Eberhard, and Cutting (1992) who used the same set of distributive referent and single referent

PAGE 66

66 vs. Although the subject head nouns of these two pre ambles are grammatically singular, they are conceptually different. In the first preamble, the subject head noun is a single token (conceptual meaning is singular), meaning that there is physically one key. But, for the second preamble, the subject head no un is a multiple token (the conceptual meaning is plural): there are multiple copies of a picture distributed across several postcards. Bock and Eberhard (1993) conducted a series of completion task experiments using a similar design to examine whether th e number information used in the agreement is notional (e.g. based on the singularity or plurality of the referent of the subject noun phrase), lexical (e.g. the grammatical number of the head of the subject NP) or morphological (e.g. the head of the subje ct NP has the plural phonological feature). Participants were provided with a set of preambles consisting of a subject head noun phrase followed by a modifying prepositional phrase in which the local noun matched or mismatched the head noun with respect t o the three kinds of numbers (notional, lexical, and morphological). Similar to the previous studies in agreement, the participants were supposed to repeat the preambles and complete the sentences with their own words. In order to test the effect of phonol ogy or morphophonology of the plural nouns, the target local nouns of Experiments 1 & 2 in Bock & Eberhard were pseudoplural nouns (singular nouns whose endings were matched to the ending of true plurals in

PAGE 67

67 their phonology) and the fillers were singular an d true plural nouns. Below are two examples: (2.4) Experiment 1: The player on the course pseudoplural / court singular / courts plural (2.5) Experiment 2: The ship for the cruise pseudoplural / crew singular / crews plural Their results showed no effect of pseudoplurals on number agreement, which indicates no evidence of phonologically conditioned agreement errors. This means that there was no tendency to produce agreement errors after singular local nouns that ended in the same sounds as plural forms (e.g. course vs. courts & cruise vs. crews ). The next question was whether it is the nominal form of plural inflection itself or the notional meaning of plurality that controls agreement. To test this, the authors examined the effect of regularity of plural noun markers. Participants repeated and completed a set of preambles with regular plural (plural in meaning and form) and irregular plural (plural in meaning but not in form) local nouns (e.g. the trap(s) for the ). The results here showed no effect of the regularity of the plural marking on agreement errors: The results for irregular plurals (e.g. foot feet ) were comparable to regular plurals (e.g. hand hands ). The results of these experiments suggest that the number agreement is not contr olled by the concrete feature of morphological marking (i.e. plural marking) but instead is controlled by some abstract specification of the number of the subject. In their final experiment, Bock and Eberhard addressed the nature of this abstract specifica tion, and in particular whether it is based on meaning or lexical stipulation or number. They used preambles with collective target local nouns to examine whether the local nouns

PAGE 68

68 cause errors that reflect their notional number (plural) or their subcategori zed number (singular). (e. g. ). When singular collectives served as local nouns, the agreement results were essentially the same as the results for the singular individual nouns. This suggests that the par ticipants treated the singular collective nouns as singular for the purpose of verb agreement. So, even for the verbs that disagreed with the subject in number, subcategorical rather than notional number predominated in the selection of verb number. From a ll of this, Bock & Eberhard (1993) concluded that phonological or semantic elements have very little influence on agreement. The control of verb number is with a lexical specification of plurality and not with plural meaning or regular plural marking. In f act, it is the subcategorized number of the subject head noun that controls the forms of the agreeing verb, not the number meaning or plural morphology. So, the results of these studies showed that subject verb agreement was influenced by attraction from the number of the local noun but they did not show any effect of animacy of the local nor any effect of length of the modifying phrase. These studies did not report any effect of plural phonology (e.g. /s/), plural morphology (regular or irregular plural n ouns), or notional number (the effect of distributive or collective nouns) in subject verb number agreement in English. Hierarchical Structure This section includes a summary of an study testing a syntactic factor (the effect of hierarchical structure of t he local noun in the subject verb agreement). Although it is about the effect of a syntactic factor, it does not reject any potential non syntactic factors. So, I consider it here as something different from the above section on work

PAGE 69

69 Bock & Cutting (1992) used a completion task in three experiments to examine the effect of length and hierarchical structure. They compared the agreement errors after phrasal (subject modifying prepositional phrase) and clausal modifiers (adv erbial clause). In their first two experiments, they used sentence fragments with complex subject NPs in which the subject head noun was followed by a prepositional phrase or by a relative clause post modifier (as in the following example). (2.6) a. b. The participants were instructed to repeat the preamble and complete the sentence with their own words. The results showed that errors were more likely following structures like (a) than (b), because in sentence (b), the local noun is separated from the subject head noun by clause boundaries. This result suggests that the local NP affects the agreement more frequently when it is in the clause of the subject head n oun than when it is embedded in a different clause. In fact, agreement errors were more frequent after phrases than after clauses that separated the verb from its head noun: The clausal m odifiers caused fewer agreement errors compared to phrasal modifi ers. In their third experiment, the length of the modifying prepositional phrase and the adverb clause were increased to test the effect of long phrases and clauses. T his experiment was a replicat ion of Experiment 1 of Bock and Miller (1991), but the conditi ons in this experiment were better controlled. Also, the material was designed for

PAGE 70

70 The result of this experiment showed that the longer phrases caused more errors while the longer clauses did not. (2.7) a. b. insulate the specification of verb agreement, so that the processes r esponsible for results support the hierarchical hypothesis. Non Syntactic Factors influencing Subject verb Production and Processing In recent years, many studies have challe nged the idea that the agreement is only a syntactic phenomenon. Many researchers have tested the effect of semantic factors such as distributive and collective nouns. Other research has tested the effect of plausibility and morphophonological influences o n agreement. In these studies, the researchers used the attraction effect of the local noun first reported by Bock and Miller (1991) to test the effect of non syntactic factors. Distributivity Vigliocco, Butterworth, & Semenza (1995) and Vigliocco, Butte rworth, & Garrett (1996a) have both examined the effect of distributive the noun is grammatically singular. Vigliocco, Butterworth, & S emenza (1995) tested the effect of semantics (i.e. the distributivity of subject head noun) in Italian. In their first experiment, they used the same methodology of Bock and Miller (1991). The preambles consisted of a subject head noun followed by a modify ing prepositional phrase. One of the variables was the distributivity of head noun, with a single token condition (e.g., the

PAGE 71

71 road to the lakes) vs. a multiple token condition (e.g., the name of the children ). The participants heard a series of sentence pr eambles which they repeated and completed with their own words. The result showed an effect of distributivity: there were more errors when the head subject was a multiple token noun. In Experiment 2 of Vigliocco et al. (1995), participants first saw an ad jective marked for singular or plural (e.g. colorato colored M,SG) then a sentence preamble (e.g. the picture(s) on the exercise book(s) distributivity (single token vs. multiple token) of the preamble. Participants were supposed to repeat the preamb le and complete the sentence using the matched adjective. The results of Experiment 2 confirmed the role of semantics (distributivity) in the subject verb agreement. These results were contrast to Bock and Miller (1991) who di d not find any effect of distributivity in English. Vigliocco, Butterworth, & Garrett (1996a) in their experiments 1 and 2, tested the effect of distributivity in Spanish. The preambles consisted of a singular/plural head noun and a singular/plural local n oun. There were two conditions: a single token condition (e.g., the road to the lakes ) vs. a multiple token condition (e.g., the name of the children ). The participants saw a series of sentence preambles which they repeated and completed. The results show ed an attraction effect for local nouns, and moreover it showed an effect of distribution of agreement errors in Spanish: there were significantly more errors in the multiple token condition than in the single token condition. In Experiment 2, Vigliocco e t al., (1996a) used the same materials of Experiment 1 in Spanish. In order to increase the error rate, they used the advantage of subject predicate Number and Gender agreement. They matched each of the preambles with a

PAGE 72

72 semantically plausible adjective. Th is is the same technique employed by Vigliocco et al. (1995) in Italian. (2.8) a. El abuelo de los nino s es viejo the.M.SG uncle.M.SG of the.M.PL child.M PL. is.3SG old.M.SG b. Los abuelo s de los nino s s on Viejo s The M.PL uncle.M PL of the.M.P child.M PL are.3PL old.M PL children In each trial, the participants saw an adjective (singular or plural) immediately followed by a sentence preamble (with singular or plural subject he ad noun and local noun) For example, they saw viejo (old. M.S ) and then El abuelo de los ninos (The .M.S uncle .M.S of the .M.P children .M.P ). They were supposed to repeat the preamble and then complete the sentence, saying El abuelo de los ninos es viejo ( The uncle of the supposed to produce the matched adjective (singular or plural) and matched verb (singular or plural). The results showed that the error rate was higher compared to Experiment 1: The error rate was higher when the adjective was not matched for the Number. The authors concluded that the high rate of errors was due to the effect of the mismatched adjectives. And finally, the error rate was higher for multiple token items (e.g. the uniform of the soldiers ) than for single token items ( the teacher for the girls ), demonstrating an effect of distributivity.

PAGE 73

73 Taken together the results of the studies (Vigliocco et al. 1995 & 1996a) conducted in Italian, Spanish, and French together demonstrated a clear effect of distributivity: more plural verbs were produced following distributive phrases than following non distributive phrases Concreteness/ I mageability of D istributive N ouns As explained above, Bock and Miller (1991) used the attraction effect to test the effect of conceptual number on agreement. In their materials, they used sentences with singular subject head nouns and plur al local nouns of a modifying prepositional phrase. (2.9) a. b. Although both of the sentence types have grammatically singular subject head nouns, the conceptual meaning of sentence (b) is plura l because it refers to multiple tokens or copies of an entity distributed across the referent of the local noun (i.e. multiple copies of a picture distributed across several postcards). They found no effect of conceptual number and concluded that conceptua l number (i.e. the distributivity of subject head noun) did not have an effect on subject verb number agreement. This result was replicated by Bock, Eberhard, and Cutting (1992) and Vigliocco, Butterworth, and Garrett (1996a; Experiment 3). Bock, Eberhard, and Cutting (1992) used the same distributive referent and single referent phrases as Bock and Miller (1991). Vigliocco, Butterworth, and Garrett (1996a; Experiment 3) used all the material of the Bock and miller (1991) except for two distributive referen t phrases. On the contrary, the studies by Vigliocco, Butterworth, & Semenza, 1995) in Italian, Vigliocco, Butterworth, and Garrett (1996a; Experiment 1 &2) in Spanish, and

PAGE 74

74 Vigliocco, Hartsuiker, Jaraema, & Kolk (1996b) in French and Dutch did show an eff ect of distributive number in subject verb number agreement. Related to these conflicting results, Eberhard (1999) investigated the effect of conceptual number of distributive subject phrase on verb agreement in English. More specifically, Eberhard tested whether the concreteness/imagability of distributive referents had any effect resembling their conceptual number. For this study, the preambles of Bock and Miller (1991) were first tested for concreteness/imagability and less concrete preambles were remove d and replaced by new concrete preambles. In the first experiment of the study, the participants saw each of the new concrete preambles of distributive referent phrases vs. single referent phrases, along with a related picture on a computer screen. Partic ipants repeated the preamble and then completed the sentence. The results showed an effect of distributivity in English sentences: The rate of agreement error was higher for the distributive subject head nouns. More plural verbs were used with the preamble s with multiple token subject head nouns than with single token ones. In Experiment 2, the same preambles were used without pictures. The participants saw only the concrete preambles, repeated them and completed the sentences. The results here showed an e ffect of distributivity, but this effect was less than the effect in Experiment 1. This difference could be due to the effect of the pictures conceptual number. In the thir d experiment, they tested the materials from Bock & Miller (1991).The participants saw each of the preambles (without pictures), repeated the preambles and

PAGE 75

75 complete the sentences. The result shows no effect of distributivity, the same as Bock and Miller ( In Experiment 4, the phrases of Bock and Miller (1991) and the new phrases used in Experiments 1 and 2 were tested together. In this experiment, participants read ding to the ease or difficulty with which it evoked a mental image of its referent. The result showed that the distributive referent preambles used in Experiments 1 and 2 were rated as easier to imagine than the distributive referent preambles taken form B ock and Miller (1991). Vigliocco, Hartsuiker, Jaraema, & Kolk (1996b; Experiments 2 & 3) were tested to determine whether the results of these experiments were due to the imageabili ty of the preambles. Similar to Experiment 4, participants read the preambles and rated each one (on a scale of 1 to 7) according to the ease or difficulty with which it evoked a mental image of its referent. The result showed that the distributive referen t preambles used in Vigliocco et al. (1996b; Experiments 2 & 3) were rated easier to imagine than the distributive referent preambles of Bock and Miller (1991). These results from Eberhard (1999) demonstrate that effects of distributivity can also be foun d in English under certain circumstances. Eberhard attributed the difference between her results and the failure to find effects of distributivity in Bock and Miller (1991) and the two replications (Bock et al., 1992; Vigliocco et al., 1996a) to difference s in the imageability of the preambles phrases. Eberhard suggests that the failure was due to the abstract nature of distributive referent subject phrases in the other studies.

PAGE 76

76 She explains that when subject phrases are highly imageable, their conceptual n umber can reliably influence the specification of number on an agreeing verb. This result indicates that the concreteness or imageability of the conceptual representation appears to affect the availability of conceptual number in the agreement process. Col lectivity Bock, Nicol & Cutting (1999) used the same procedure as Bock and Miller (1991) to assess whether number agreement is based on notional meaning (conceptual meaning) or grammatical meaning (form based). Using a completion task, the number agreement in verbs, reflexive pronouns and tag questions was examined under controlled conditions. The head of the subject noun phrase could be either singular, plural, or a collective noun (nouns which are notionally plural but grammatically singular). Groups of p articipants were instructed to repeat the preamble and complete the sentence with a verb or a reflexive pronoun or repeat the complete sentence and add a tag question to it as shown (underlined) in the following examples: (2.10) a. Verb group: the actor(s )/cast in the soap opera(s) was/were popular b. Reflexive pronoun: the actor(s)/cast in the soap opera(s) watched himself/ themselves. c. Tag question: the actor(s)/cast in the soap opera(s) rehearsed he/they? The results showed an effect of col lectivity on subject verb agreement. In their followed by plural noun in the post modifying phrase, elicited 60% plural verbs. For ordinary singular head nouns (again with plural noun in the post modifying phra se), this rate was 10%.

PAGE 77

77 It was also found that in sentences with collective head nouns, verbs were more sensitive to the grammatical number while pronouns were more sensitive to the notional ber under control language production. Gender Agreement ( C onceptual I nformation) Vigliocc o and Franck (2001) differentiate between syntactic information (for conceptual information (referring to a male or female entity). Syntactic information is considered first source of information and conceptual inf ormation is the second. In two completion experiments (1 & 2) in Italian and French, participants were provided with a set of texts introducing a male or female actor (conceptual gender) in preambles composed of a subject head noun, a local noun and predi cative adjectives. Participants read each introduction silently, then repeat the preamble and completed the sentence using the adjective. The authors manipulated the congruency between the conceptual gender of a referent introduced in the introduction sent ence and the grammatical gender of the noun with which the adjective had to agree. They also manipulated gender match/mismatch between the subject head noun and the local noun. The results showed that errors were more common when the gender of the referent and the gender of the noun were incongruent. The authors concluded that both conceptual and syntactic information were taken into account during grammatical encoding.

PAGE 78

78 In two further completion experiments (3 and 4) in Italian and French, the participants were presented with the same preambles and predicative adjectives, but the introduction texts were not used (to avoid any effect of conceptual information). The authors manipulated the gender match/mismatch between the subject head noun and the local noun. The results showed a gender mismatch effect. Moreover, comparing the results of the context and no context experiments indicates that having an incongruent context hurts performance (i.e., creates agreement errors) whereas having a congruent context has s ome benefits. Thus, working on gender agreement in Italian and French, Vigliocco and Franck (1999, 2001) found differing agreement behavior depending on whether conceptual information was consistent with linguistic gender, conflicted with linguistic gender or was neutral. Discussing maximalist positions with respect to gender agreement, Vigliocco and Frank (2001:370) mention that the Maximalist approach argues that both the grammatical gender of a noun and the conceptual information are used to maintain a ccuracy. If the grammatical gender of the noun and the conceptual information are compatible, the conceptual information reinforces the agreement process. But, conceptual information poses problems if that information is different from the grammatical gend er (i.e., the syntactic information). In this case, there is a competition between the two kinds of information and it makes speech production more difficult. Morphology Vigliocco, Butterworth, and Semenza (1995) compared rates of agreement errors for inv ariant plurals and ordinary plurals in Italian. Invariant plurals occur when the singular and plural forms of a noun are the same, like English fish and sheep For these nouns, morphophonology does not show the number (singular and plural forms are the

PAGE 79

79 sam e). In Italian, the nouns were presented along with a number marked determiner, so that the number of the noun was never ambiguous (as per the following examples). (2.11) ordinary plural: il gatto sui tetti i gatti sul tetto the cat on the roofs the cats on the roof (2.12) invariant plural: la citta sulle colline le citta sulla collina the sg town on the hills the pl town on the hill The participants were supposed to repeat the preamble and complete the sentence. The result showed that more errors occurred with the invariant nouns than with ordinary nouns. Plausibility Thornton and MacDonald (2003) tested the effect of noun verb plausibility effects on agreement production. In each trial of their Experiment 1, the participants were presented with the past participle of a verb (e.g., played or praised ) displayed at a computer screen followed by the auditory presentation of a preamble (e.g., the album by ). In Experiment 2, the participants heard the preamble then saw the past participle of the verb. In both experiments, participants were instructed to create a complete passive sentence (e.g., the album by the classical composers was played ). The plausibility relations were manipulated via the verb. For one set of verb s both the subject head noun and local noun of the preamble were plausible but for the other set of the verbs only the subject head noun was plausible and semantically related to the agreeing verb. For example, the subject head noun ( album ) was always a pl ausible object of the verb (e.g., album can be played or praised ) but the local noun ( composers )

PAGE 80

80 varied in its relationship to the verb (e.g., composers can plausibly be praised but not played ). The results of both experiments showed that the participants produced more agreement errors when the past participle could plausibly apply to the local noun (e.g., composers can be praised ) than when it could not ( composers cannot be played ). The authors concluded that agreement production is sensitive to plausibil ity effect. The plausibility of subject head noun affected the production of subject verb agreement. in other words, intervention of a plausible (rather than implausible) local noun between the subject head noun and its verb causes more agreement error. Fi nally, Thornton & MacDonald (2003) discussed that manipulation of plausibility involve semantic (non syntactic) information outside of the subject noun phrase. Comprehension of A greement In the literature of the production and comprehension of agreement, t he absence of semantic effects was taken as initial evidence for an independent, encapsulated stage of syntactic processing ( e.g. the minimalist approach). But, this claim was challenged by researchers who showed evidence of effects of non syntactic factor s in the production of subject verb agreement. Regarding the effect of non syntactic factors in the comprehension of subject verb agreement, Vigliocco and Hartsuiker (2002) refer to influential research (e.g., MacDonald, Pearlmutter, & Seidenberg, 1994; T abor & Tanenhaus, 1999; Tanenhaus & Trueswell, 1995) providing evidence indicating that comprehenders use all available information as the speech signal unfolds (including phonological, metrical, syntactic, and crucially semantic information). The Constrai nt based framework (MacDonald, Pearlmutter, & Seidenberg, 1994; Trueswell, Tanenhaus & Garnsey, 1994) suggests

PAGE 81

81 that multiple probabilistic sources of information are integrated at the early stage of comprehension. Following Bock and Miller (1991), many st udies used completion tasks in an elicitation paradigm to test the effect of attraction, hierarchy and other non syntactic factors in the production of subject verb agreement. Evidence from agreement processing in comprehension can be a useful complement t o the evidence from agreement production studies. Wagers, Lau, and Phillips (2009) explain that the results of studies in comprehension of agreement are congruent with the evidence from studies on production In production studies, researchers measure the rate of agreement errors that the speakers produce to test the effect of different factors. Comprehension studies use acceptability judgments (e.g., Clifton, Fraizer, & Deevy, 1999; Haussler & Bader, 2009) or processing difficulty as reflected either in di fferences of reading time s(e.g., Pearlmutter, Garnsey, & Bock 1999), or in differents event related brain potentials (ERPs) (e.g., Kaan, 2002). Pearlmuter, Garnsey and Bock (1999) used three experiments to examine the process of subject verb agreement in sentence comprehension. In Experiment 1, they used a word by word self paced moving window reading methodology in which the participants were supposed to read sentences (e.g., the key to the cabinet(s) was/were rusty from many years of disuse) The resul ts showed that reading times after the verb were faster if the head noun, the local noun, and the verb were all singular. The reading time after the verb was slower, if either the local noun or the verb was plural. The eye tracking paradigm in Experiment 2 showed a similar result: an effect of agreement computations was observed on the word following the verb. Also, processing

PAGE 82

82 disruptions were observed in both agreement violations and locally distracting number marked nouns. Experiment 3 showed an asymmetr y in the pattern of disruptions that is in line with error distributions in language production (e.g., Bock & Miller, 1991). These results indicated that comprehenders process agreement errors more easily exactly in those situations in which speakers prod uce more agreement errors. Pearlmutter, Garnsey, & Bock (1999) concluded that agreement is an early, integral component of comprehension, mediated by processes similar to those of production. Thornton and MacDonald (2003; Experiment 3) also tested the effe ct of plausibility in the comprehension of agreement. They used an error elicitation paradigm in a word by word self paced reading comprehension task and manipulated the extent to which the head and local nouns were plausibly related to the agreeing verb. Participants were provided with a complete sentence consisting of a head noun, a local t he feeling(s) about the undergraduate student(s) was noticed/shared by the dean ). There were ei ght conditions depending on the singularity vs. plurality of the head noun and the local noun and also the plausibility of the local noun. The head noun ( feeling ) was always a plausible object of the verb (e.g., the feeling can be shared or noticed ) but th e local noun ( undergraduate student ) varied in its relationship to the verb (e.g., s tudent can plausibly be noticed but not shared ). times at the verb and surrounding words were analyzed. The result showed strong effects of plausibility on the comprehension of subject verb agreement. There were longer reading times at the verb when both the subject head noun and the local noun

PAGE 83

83 were plausible than in the condition where only t he head subject noun was plausible. This indicates that processing was more difficult when both of the noun phrases were plausible than when only the head subject noun was plausible This result is in line with the production experiments (Thornton & MacDon ald, 2003; Experiments 1 and 2).

PAGE 84

84 to output systems Figure 2 1 Hierarchical encapsulated model by Bock and Levelt (1994) M ESSAGE FUNCTIONAL PROCESSING Lexical Selection Function Assignment POSITIONAL PROCESSING Constituent Assembly Inflection PHONOLOGICAL ENCODING

PAGE 85

85 Figure 2 2 Two accounts of the relationship between n umerosity

PAGE 86

86 Figure 2 3 Marking and Morphing model of agreement for verb agreement and attraction

PAGE 87

87 CHAPTER 3 EXPERIMENT 1 JUDGMENT TASK Overview As discussed in Chapters 1 and 2, there is subject verb agreement in Persian (for person and number), b ut this agreement is optional when the subject of the sentence is an inanimate plural noun. In this case the verb of the sentence may be marked for third person singular or plural and both of these forms are acceptable and grammatical. Lotfi (2006:136 137) explained the potential role of verb type, tense and aspect discussed in Chapter 1, Lotfi argues that Persian speakers prefer to use plural verbs when the sentence is in t he future tense. Further, Lotfi predicts that verb type also has an important impact on agreement systematic experiments investigating the effect of these factors. However, he suggests that it may be more ac ceptable to have autonomy in actions when the course of events has not finished yet. Saasat (1996) shows a series of examples from literature texts showing that verbs of inanimate nou ns in classic texts were singular, but that in a few exceptional exampl es the inanimate plural nouns had plural verbs. In these cases, he argues that the inanimate nouns can have plural verbs because they are used metaphorically (e.g. personified as living creatures).

PAGE 88

88 Goals Acceptability G oals This experiment had two sets of goals. The first set of goals (which I call existence of optional subject verb agreement in Persian. It aimed to test whether the Persian comprehenders actually accept se ntences with mismatched agreement (i.e., the inanimate plural subjects have singular verbs), and additionally to test the effect of Verb Tense and Verb Type on the acceptability of optional agreement by Persian comprehenders. In other words, this experimen t tested whether the tense of the sentence (present vs. past) causes comprehenders to accept a higher proportion of sentences with a mismatched (i.e. singular) verb. In a pilot study, the effect of future vs. past was tested and the results showed that pa rticipants we re not using mismatched verb forms with future tense. The reason for this could be the infrequent use of future tense verbs in informal spoken Persian; Persian speakers prefer to use ordinary present tense (see past and present tense in Persia n, Chapter 3) to talk about future events. So, the effect of verb tense was tested here using present vs. past tense. mismatched verbs would change as a function of the type of verb used. It is worth noting here that the materials of this experiment were the complete sentence versions of the target preambles used in Experiment 2, and so this experiment also serves to confirm the acceptability of those preambles in Experiment 2.

PAGE 89

89 Role Interpretation Goals Related to the issues of verb type, a secondary goal of this experiment was to examine the relationship between the acceptability of a mismatched verb form with the dissertation). Essentially, I wanted to test whether there is any relationship between the degrees of acceptability for singular verb forms and how readers interpret (perceive) the role of the corresponding inanimate plural subject noun. For this purpose, I analyzed the an effect of Verb Tense (present vs. past), 2) the effect of Verb Type (eight types of verbs to be explained in the Material section of this chapter), a nd 3) the effect of verb Thus, in this experiment, participants gave two types of responses after each of a series of complete sentences that were either in past or present tense: First, they read the sentence and then judged whether it was acceptable in Persian or not. Second, they provided a response about whether they judged the subject of the sentence as a "do er", something that was "acted on", or neither of thes e. Since the thematic role of the subject nouns in some of the verb types such as group C (verbs of emission), group D interpretation for the role of the subject noun, in case they were thinking that the role of

PAGE 90

90 Method Participants The participants were 29 Iranian native speakers of Persian (15 males and 14 females) between 22 30 years old (mean 26 years). They were members of the university of Florida community, and participated in the study vo luntarily. The participants were late learners of English who had learned this language for the purpose of education and did not have significant knowledge of any other language. All participants had normal or corrected vision and did not have any reported cognitive or neurological impairment. 25 of the participants in this experiment participated in Experiments 3 (the effect of unity) and 4 (the effect of concreteness), which were conducted together and more than one year after this experiment. Materials A set of one hundred and twelve target sentences in 4 conditions (past vs. present, and singular vs. plural verbs) were created, divided across 8 verbs types, with 14 verbs of each type. Preparing a list of potential sentences with inanimate plural subject nouns, I classifie d the sentences into different verb type groups in order to systematically classify them according to their structure and the potential thematic roles that the verbs would assign to the subject of these verbs. The target sentences consist ed of an inanimate plural subject noun followed by a combined compound verb. As discussed in Chapter 1, a combined compound verb consists of a non verbal constituent plus a verbal part. I will next describe the salient features of tense in Persian, and th e types of verb that were used in the study.

PAGE 91

91 Past and p resent t ense in Persian The verbs used in this experiment were either present or past tense. Both of the verbs were indicative and positive. There are two types of indicative present tense verbs in Pe rsian: ordinary present and progressive present. The present tense verbs used in tense form i n Persian used only for generic statements. The ordinary present is formed by the indicative prefix /mi / + present stem + verb ending, as seen in (3.1) below. (3.1) a. gol ha ro mi chin e flower PL OM IND pick 3SG b. hr ruz gol ha ro mi chin e every day flower PL OM IND pick 3SG Although, in Persian progressive (continuous) aspect is shown by the auxiliary inary present tense is commonly used to convey the continuous aspect of the action (for this reason, some linguists gloss the prefix /mi / as a continuous/ durative marker) and the present progressive verb is used when speakers want to emphasize the contin uity of the action. (3.1) c. gol ha ro mi chin e flower PL OM IND pick 3SG

PAGE 92

92 d. gol ha ro dar e mi chin e flower PL OM have 3SG IND pick 3SG So, the ordinary present verb ca n convey the meaning of continuous or incontinuous ordinary present or both of them in one sentence (as the existence of one does not contradicts the other; a person can do a durative action every day). (3.1) e. hr ruz gol ha ro mi chin e every day flower PL OM IND pick 3SG f. hr ruz gol ha ro dar e mi chin e every day flower PL OM have 3SG IND pick 3SG Ordinary present can also be used to convey that events took place in the future. In this case, an adverb of future can be added, if needed (Mahootian 1997: 238). Below is a set of examples: (3.2) a. b rf mi bar e snow IND fall 3SG

PAGE 93

93 b. frda brf mi bar e tomorrow snow IND fall 3SG c. h r ruz b rf mi bar e every day snow IND fall 3SG So the present tense verbs in this experime nt could have habitual, continuous, or ordinary present meaning. There are four types of indicative past tense verbs in Persian. Simple past (formed by past stem + verb endings) which conveys the perfective aspect of an action that has been done in the pa st. This verb tense implies the completion of an action in the past without emphasizing on the duration of the action. The past tense verbs used in this experiment were of this type. (3.3) a. gol ha ro chid flower PL OM pick.Past.3SG The ordinary past tense (formed by the prefix /mi / + past stem + verb endings) conveys the imperfective aspect of an action in the past. This verb tense my be used to convey progressive aspect, too.

PAGE 94

94 (3.3) b. gol ha ro mi chid flower PL OM IN D pick.Past.3SG The progressive past tense (formed by the verb /dasht endings + the indicative prefix /mi / + past stem + verb endings) specifically emphasizes the imperfect aspect and continuity of t he action. (3.3) c. gol ha ro dasht mi chid flower PL OM had.3SG IND pick.Past.3SG The fourth type of indicative past tense verb in Persian is Past Perfect (formed by past participle form of the main verb + past te verb tense conveys that an action had taken place before another action in the past. It specifically emphasizes the perfective aspect of the verb. (3.3) d. gol ha ro chide bud ke Niki um d flower PL OM pic ked.P be.Past.3SG that Niki came.3SG As was mentioned above, the verbs in this experiment were ordinary present tense vs. simple past tense. The ordinary present tense verbs have imperfective aspect and can ha ve progressive me aning ( although they are not progressive). As was mentioned above, both of the present tense verbs (ordinary present and present progressive) imply continuity in the verb (so I did not have any choice for choosing a present verb in Persian with out continuity). For the list of the sentences with present

PAGE 95

95 tense verbs, the participants were told to imagine that the events happen on the same day. On the contrary, the simple past tense verbs have perfective aspect but do not have progressive me aning. For the past tense verbs, the participants were told to imagine that the events happened the day before. I chose ordinary present vs. simple past tense verbs because ordinary present has imperfective aspect but simple past has perfective aspect. In fact, I selected these two types of verbs to test whether the completion of an action (perfective vs. imperfective aspect of a verb) has any effect on the optional subject verb agreement or not. Verb t ypes Thinking about the types of verbs that inanimate subject nouns can have and the effect of different kinds of verbs (with different structures) led me to use several different types of verbs that vary primarily with respect to how they assign thematic roles to their subject. Eight different types of verb s were used, leading to variation in the structure of the sentence and the thematic role the verbs assigned to the subject nouns (Levin, 1993 & Dabir Moghaddam 1982 & 1997). The verb types and the thematic role of the subjects are explained below: Group A Unaccusative verbs These are the noun + verb combined compound verbs explained in chapter 1 (subject agent) NP (object patient) has lost its NP (subject agent) So, the NP (object patient) has taken the position of the subject (valency reduction). Perlmutter (1978) has characterized the unaccusative verb as the verbs that take an internal argument but no external argument. Levin (

PAGE 96

96 unaccusative verbs and passive verbs are the same as they have no external argument but a direct internal argument. Levin (1993) explains that the verbs in the reduced valency structure are unaccusative and the su bject of the sentence is in fact the underlying object with patient thematic role. The example below shows a transitive structure (a) and its related unaccusative structure (b). (3.4) a. Niki tnab ha ra gere zd Niki rope Pl OM knot strike.Past.3SG b. tnab ha gere xord n rope Pl knot collide.Past 3PL So, the target sentences of this judgment task included 14 sentences with unaccusative structure in four conditio ns (past vs. present and singular vs. plural). These verbs are classified as group A. (3.5) a. tnab ha gere mi xor e /mi xor n rope Pl knot IND collide 3SG /IND collide 3PL

PAGE 97

97 b. (tnab ha gere xord /xord n rope Pl knot collide.Past.3PL /collide.Past 3PL Group B Entity specific inchoative verbs These are the adverb + verb or noun + verb combined compound verbs explained in Chapter to the entities t verbs used in this group is: sprout, flower, bud (produce leaf), grow, subside rise, swell, s cab over, blis ter, rust, blossom, bud, mold, and take root These verbs are inchoative verbs which convey entity specific changes of state. In this research, they are referred nditions: (3.6) a. lubiya ha jvane mi zn e /mi zn n bean Pl sprout IND strike 3SG /IND strike 3PL b. lubiya ha jvane zd /zd n bean Pl sprout strike.past.3SG /strike.past 3PL

PAGE 98

98 Group C Verbs of emission These are the noun + combined compound verbs explained in Chapter 1. specific modes of bei the verbs are inanimate and the verbs can be classified in different categories. The verbs of emission used in this experiment were 1) verbs of light emission (e.g light, flicker, gleam, spark, and flash ) 2) verbs of sound emission (e .g. rustle jingle, moan, buzz, make noise, and rattle ) 3) verbs of substance emission (e.g. bubble, gush, and ooze ). The subjects for these verb have a theme role. (3.7) a. pnjre ha tq tq mi kon e /mi kon n window pl rattling IND do 3SG /IND do 3PL b. pnjre ha tq tq krd /krd n window pl rattling do.Past.3SG /do.Past 3PL Group D Verbs of i nstrument These are the Adverb + verb or noun + verb combined compound verbs (subject agent) NP (o bject patient) NP (oblique instrument) (subject agent) and NP (object patient) So, the NP (oblique instrument) has taken the position of the subject (valency reduction). Referring to other studies, Levin (1993:274)

PAGE 99

99 refers to this NP i instrument NP can promote to subject position or not depends on the verb and the instrument. The first be promoted. (3.8) a. Niki nushabe ro ba nei xord Niki soda OM with straw drank.3SG b. *nei nushabe ro xord Straw soda OM drank.3SG *The straw drank the soda. (3.9) a. Niki xbr ro ba bolndgu ha elam krd Niki news OM with loudspeaker Pl announce did.3SG b. boln dgu ha xbr ro elam krd n loudspeakers news OM announced The target sentences included 14 sentences of this group (group D). They are intransitive sentences with intermediary instrument as subje ct of the sentence(again in

PAGE 100

100 four conditions). These verbs are intransitive with instrument subjects. In this (3.10) a. helikupter ha chrx mi zn e /mi zn n helicopter pl rotating IND strike 3SG /IND strike 3PL b. helikupter ha chrx zd /zd n helicopter pl rotating strike.Past 3SG /strike.Past 3PL Group E Verbs of a gency These are the noun + verb combined compound verbs explained in Chapter 1. Because of the nature of the actions, the verbs of this group require animate subjects with agent thematic role. Normally, inanimate subjects cannot perform these verbs unless in the hypothetical environment such as cartoons and fiction stories. These type of sentences were tested to investigate whether the mismatched form of the verb is used in case the subjects are used metaphorically and personified as human bei ng (3.11) a. tr ha gerye mi kon e /mi kon n perfume Pl cry IND do 3SG /IND do 3PL

PAGE 101

101 b. tr ha gerye krd /krd n perfume Pl cry did.3SG /did 3PL Group F Passive verbs The structure of the passive sentences is NP (patient) + past participle + passive auxiliary (as explained in Chapter 1). The subject has patient thematic role. Th e verb is transitive, the original agent subject is deleted and the object has been promoted to Below is a set of target sentence in four conditions: (3.12) a. golab i ha xorde mi sh e / mi sh n pear Pl eaten IND become 3SG /IND become 3Pl b. golabi ha xorde shod /shod n pear Pl eaten became.3SG /became 3Pl Group G Stative a As explained in Chapter 1, in combined compound verbs (Adjective + Auxiliary), this experiment, the present form of the number and person (clitic / that

PAGE 102

102 either /bud/ (be.past.3SG) or /bud n 3PL). The thematic role of the subject could be characterized as none (not patient, not agent). Below is a set of target sentence in four conditions: (3.13) a. miz ha chubi ye / yn table Pl wooden be.3SG /be.3PL b. miz ha chubi bud /bud n table Pl wooden be.Past.3SG /be.Past.3SG Group H Inchoative As was explained in Chapter 1, under combined compound verbs (Adjective + Levin (1993:244 245) provides a long list of this kind of verbs. This group of sentences he subject. The thematic role of the subject is Patient. Below is an example of a target sentence in 4 versions: (3.14) a. rng ha xoshk mi sh e /mi sh n paint pl dry IND become.3SG /IND become 3PL

PAGE 103

103 b rng ha xoshk shod /shod n paint pl dry became.3PL /became 3PL This group is similar to group B, as both of them include change of state. But they are classified in separate groups as the change of state in this group is not inherited and not limited to the subject nouns. Also, they have different structure from the sentences in group (B). The structure of this group includes: subject + adjective + tructure of group (b) includes subject + noun/adverb + verb. The distracter sentences were a set of 80 sentences with different structures in two conditions/lists (past and present). The filler preambles included different kinds of structures. They consist ed of a singular or plural (in)animate subject noun, optional singular/plural (in)animate direct/indirect object noun, the direct object marker /ra/ (if there was direct object), and the nonverbal component of a (in)transitive combined compound. Using a L atin square design, the target sentences were divided into four compatible experimental lists (two lists in past and two in present tense). In each list, there were 112 target sentences (56 sentences with plural verb and 56 sentences with singular verb). T here were 14 sentences of each verb type group. So, manipulating the Tense (present vs. past) and Number of the verb (singular vs. plural), the target items included 4 types of sentences (for each of the sentences in each group): Singular present tense ve rb Singular past tense verb

PAGE 104

104 Plural present tense verb Plural past tense verb The eighty filler preambles were the same but the tense of the verbs was present or past. The past vs. present tense filler sentences were added to the lists of past vs. present t ense target sentences so that the participants saw only present or past sentences in a single list. The order of the trials was arranged pseudo randomly so that there were no subsequent target sentences with the same condition. Each list was read and judge d by fourteen participants. A list of the target and filler items in gloss and in Persian language along with the related condition are available in Appendices A at the end of the dissertation. Since understanding the literal meaning (the third line of the gloss) could be difficult for non native speakers of Persian, an English grammatical equivale nt of the complete sentence was added in the forth line. Procedure A questionnaire methodology was used. Participants were first asked to read and sign a consent form for voluntarily participation in the experiment. The task started with a training section that included instructions and four practice trials. When the participants showed their readiness, the real experiment began. In order to test the effect of tens e, the experiment was done in two sections (with two compatible lists). In each section, the participants received a list of the complete sentences (56 target and 40 filler items). In one of the sections, the tense of the list of sentences was present whil e in the other section the list of sentences was past. There was no adverb of time in the sentences, and so participants were told to imagine that all the events of the sentences happened the day before (past) while the event of the other list happened on

PAGE 105

105 the same day (present). In both of the lists, half of the target sentences had singular verbs and the other half had plural verbs. So, half of the participants read and completed the list (A) with the past tense verbs and the list (B) with the present tens e verbs and vice versa. As noted above, participants did two things with each sentence. First, they circled / qabel e qbul / / qeir e qabel e qbul / e ml/, acted on), on a line next to the sentence. The entire session, including all written materials and conversation, was conducted in Persian. Design and Data Analysis Acceptability For the first goal of the experiment (acceptability), the part were entered into an Excel spreadsheet, coded zero or one based on whether the participants accepted (0) or rejected (1) the sentence. Recall that the question for this experiment is whether there was 1) an effect of Verb Tense (present vs. past) and 2) the effect of Verb Type in the preference for optional subject verb agreement. The dependent variable was the proportion of rejected mismatched (singular) verbs out of all items and there were two independent, within subject factors: ver b tense (with two levels, present and past) and verb type (with eight levels, A H listed above). Sets of by participants (F1) and by items (F2) Repeated Measures ANOVAs with the two level factor Verb Tense and eight level factor Verb Type performed. In ad dition, planned paired sample t tests were performed. An alpha level of .05 was used for all statistical tests.

PAGE 106

106 Role i nterpretation For the second goal of the experiment (role interpretation), responses were entered into an Excel spread sheet, coded zero or one based on the Rec all that the question for this experiment is whether there was 1) an effect of Verb Tense (present vs. past) and 2) an effect of Verb Type, and 3) an effect of Verb noun nd So, there were three independent, within subject factors: verb tense (with two levels, present and past) and verb type (with eight levels, A H listed above), and verb number (with two levels, singular and plural). Sets of by participants (F1) and by items (F2) Repeated Measures ANOVAs with the two level factor Verb Tense, eight level factor Verb Type, and two level factor Verb Number were performed. In addition, planned paired sample t tests were performed. An alpha level of .05 was used for all statistical tests. participant was using formal language rather than informal spoken Persian. The

PAGE 107

107 Ques tions of the Experiment Acceptability (1) Does the tense of the verb (present vs. past) have any effect in the comprehension of subject verb agreement? More specifically, does the present vs. past form of the verb cause Persian native speakers to accept gr ammatical number mismatches? (2) Does the type of the verb (type A H above) have any effect on the comprehension of subject verb agreement? More specifically, does the type of the verb cause Persian native speakers to accept the mismatched form of the verb (singular) more often with inanimate plural subjects? Role i nterpretation (1) Does the tense of the verb (present vs. past) have any effect in the the present vs. pa st tense of the verb cause Persian native speakers to interpret different roles for the inanimate plural subject nouns in sentences with different verb types (type A H) and with matched (plural) or mismatched (singular) verbs? (2) Does the type of the verb (type A thematic role of he subject nouns? More specifically, does the type of the verb cause Persian native speakers to interpret different roles for the inanimate plural subject nouns in sentences with present or past tense and matched (plural) or mismatched (singular) verbs? (3) Does the number feature of the verb (singular vs. plural) have any effect on the does the s ingularity or plurality of the verb cause Persian native speakers to interpret different roles for the inanimate plural subject nouns in sentences with different verb types (type A H) in present or past tense?

PAGE 108

108 Predictions Acceptability Regarding the effect informal pilot study, I predicted that more mismatched (singular) form of the verbs would be accepted in the sentences with past tense compared to sentences with present tense. For the effect of ver b type, participants were predicted to accept more singular verbs in A (unaccusative verbs), F(passive verbs), G (stative auxiliary verb), H (inchoative auxiliary verb) than in B (entity specific inchoative verbs), C (Verbs of emission), D (verbs of instru ments), and E (verbs of agency). As the subjects in groups A, F, and H receive patient thematic role (and group G with no thematic role) from their related verbs while the subjects in B, C, D and E do not. Role i nterpretation For the effect of Verb Tense, the participants were predicted to interpret more rather than present. For the effect of number feature of the verb, the participants were d with singular verbs rather than plural, as the singularity of the verb may imply that the subject is acted on. For the effect of verb type, participants were predicted to interpret e for the inanimate subject nouns in sentences with A (unaccusative verbs), F(passive verbs), H (inchoative auxiliary verb) than in G (stative auxiliary verb), B (entity specific inchoative verbs), C (verbs of emission), D (verbs of instruments), and E (ve

PAGE 109

109 Results Result of a cceptability of sentences were entered into sentences. The results showed that virtually none of the sentences with plural verbs (either with past or present tense) were rejected (wit h a rejection rate of only 0.001). This means that the participants accepted target sentences with matched verbs (plural) belonging to any verb group with present tense or past tense. Next, I analyzed the rejection rates of sentences with mismatched verbs (singular) to test the effect of Verb Tense and Verb Type on the acceptability of the mismatched verbs. There were two independent, within subject factors: Verb Tense (with two levels, present vs. past) and Verb Type(with eight levels, groups A H). This a nalysis revealed a main effect of Verb Type (F1(7, 189)=353.047, MSE=0.024, P<0.01). There was no effect of Verb Tense(F1=2.420, F2=1.269) or an interaction between Verb Type and Verb Tense (F1=1.217, F2<1). The results are reflected in T ables 3 1 and 3 2. These results show that the Verb Type had a significant effect on the Verb Tense nor the interaction between Verb Type and Verb Tense had an effect on the acceptance of the mismatched form. These results indicate that Verb Type (group A H) was a significant factor in the acceptability of optional subject verb agreement in Persian. Table 3 3 shows the mean nacceptability of the singular form of the verb, broken down by verb type.

PAGE 110

110 As Table 3 3 indicates, the participants showed a similar range of rejection rates for the mismatched verbs in unaccusative verbs (groups A, 0.04), entity specific inchoative verbs (group B, 0.07), passive verbs (group F, 0.02), verb of stative auxiliary groups (A, B, F, G, H), the mean proportion of rejection of mismatched verbs is below 0.1. This means that participants accepted between 93% to 97% of the target sentences with inanimate plural subject nouns and singular verbs. Considering the thematic role of the subject nouns in these types of the verbs; patient or patient like (patient for ve rb groups A, F,& H, theme for verb group B, and none for the verb group G), we may conclude that the Verb Type and the thematic role the verb assign to its subject has a crucial role in optional subject verb agreement. The rejection rates for mismatched v rejected 96%) are much higher than any other verb groups. This means that the participants accepted about 76% of the target sen tences with inanimate p lural subject nouns and singular verbs in group (C), 20% in group (D), and only 4% in group (E). Considering the thematic role the verb groups assign to their subject nouns, we may conclude that the Verb Type and the thematic role th e verb assign to its subject has an important role in the optional subject verb agreement. The verbs of agency assign what thematic role the verbs of emission as sign to their subject noun because the subjects are inanimate nouns and do not have voluntary control on the action.

PAGE 111

111 s from the agent verb (group E). E) is .04, meani ng that participants did not accept the mismatched (singular) form of the verb for the animate verbs (e.g. cry, dance, walk). In fact, Persian comprehenders were expected to reject all the sentences with the mismatched verbs. This small acceptance proporti on (0.04) is because of the stereotypical meaning of one of the target sentences explained below: (3.15) baqlva ha cheshkk mi zn e /mi zn n baklavas wink IND strike 3SG /IND strike 3PL The is for this reason that some of the participants accepted the mismatched form of the Table 3 4 shows the pairwise comparison of the mean proportion of the comparison of the mean proportion of acceptability shows, the verbs in groups (A, B, F, G, and H) do not show mean difference with each other while they show mean difference with group (C, D, and E). But the verbs in groups (C, D, and E) show mean

PAGE 112

112 difference with each other and with other verbs in groups (A, B, F, G, and H ). The calculation is based on estimated marginal means and adjustment to multiple comparisons: Bonferroni. The Mean difference is significant at the 0.05 level. Result of r ole i nterpretation sentences were entered into an Excel spreadsheet, c patient interpretations. There were three independent, within subject factors: Verb Tense (with two levels, present vs. p ast), Verb Num ber (with two levels, singular vs. plural), and Verb Type (with eight levels, groups A analysis, as most of the participants interp subjects do not have agent or patient role. So, there remained only seven verb types (verb groups: A, B, C, D, E, F, & H) explained The analysis revealed a main effect of Verb Type (F1(6, 126)=549.120, MSE=0.033, P<0.01) and main effect of Verb Number (F1(1, 21)=4.745, MSE= 0.26, P<0.05; F2(1, 90)=4,523, MSE= 0.015, P<0.05). But, the results did not show a n effect of Verb Tense, by participants (F1= 2.839), although it did reach significance by items (F2 (1, 90)=6.535, MSE= 0.012, P<0.05). There was no interaction between Verb Type and Verb Tense by participants (F1 < 1, although it did reach significance by items (F2 (6, 90)=2,257, MSE=0.012, P< 0.05). There was no interaction between Verb Type and Verb Number (F < 1, F2 < 1), no interaction between Verb Tense and Verb Number (F < 1, F2< 1), and no interaction

PAGE 113

113 among Verb Type, Verb Tense, and Verb Number ( F < 1, F2 < 1). The results are shown in Tables 3 5 and 3 6. These results show that Verb Type (group A, B, C, D, E, F, & H) had a significant sentences. Table 3 7 the patient thematic role of the subject nouns of the sentences with the verbs of each group. As Table 3 7 indicates, for the subject nouns with unaccusative verbs (verb group A), participa nts had a patient interpretation rate 0.67 while the inchoative verbs (verb group B) elicited only a 0.18 rate. These results are interesting, as they show the The participa interpretation rates for the verbs of emission (group C) and verbs of instrument (group D) were both similarly low (0.05 & 0.01), and for the subjects of verbs of agency (group E) the proportion is zero. This means that the par ticipants did no t consistent with the linguistic agent role of the subject nouns of the sentences of that group. Finally, the participants showed a similar rate of patient interpretation for the Grammatically, the subjects of sentences with these two verb groups have acted on (patient) role. These figures show that the ratings substantially agree with l inguistic analyses and intuitions about these verbs.

PAGE 114

114 Table 3 8 shows the pairwise comparison of the mean proportion of the on interpretation of the role of the subject nouns for each verb group. As the comparison of the mean proportion of acted on role shows, the verbs in groups (F & H) do not show mean difference with each other while they show mean difference with the rest of the groups. The verb groups (C & D) and (D & E) show similarities with each other and differences with the res t of the groups. The verb groups (A & B) do not show any similarity with each other and with any other verb group. The calculation is based on estimated marginal means and adjustment to multiple comparisons (Bonferroni). The mean difference is significant at the 0.05 level. The results showed that the Verb Tense (present vs. past) did not have a nouns of the sentence. Table 3 9 shows the mean proportion of the partic on role interpretations for the subject nouns of the sentences with present and past tense verbs. Table 3 10 shows the pairwise comparison of the patient role interpretation in sentences with present and past tense verbs. This table shows th at the interpreted thematic roles in the sentences with present or past tense verbs were not significantly different. This means that the present or past tense verb did not cause the participants to consider the subject nouns as acted on. The results showe interpretation of the thematic role of the subject nouns. Table 3 11 shows that the participants perceived more patient role for the subject nouns of the sentences in which the verbs were mismatche d (singular) rather than matched (plural). This means that the

PAGE 115

115 mismatched (singular) form of the verbs caused the participants to interpret the subject nouns more as acted on. Table 3 12 provides the pairwise comparison of the patient interpretation in sen tences with singular vs. plural verbs. This table shows that the proportions of the interpreted patient roles in sentences with singular or plural verbs were different. Finally, Table 3 13 shows the proportion of patient interpretation for the inanimate pl ural subject nouns in sentences with different verb groups and singular or plural verbs. Discussion This experiment had two sets of goals. The acceptability goal was designed to experimentally confirm and quantify the existence of optional subject verb ag reement in Persian and to test the effect of Verb Tense and Verb Type on the acceptability of optional agreement by Persian comprehenders. The Role Interpretation goal was designed to examine the relationship between the acceptability of a mismatched verb form with the perceived thematic role of the nouns of the sentences with three factors; the effect of Verb Tense, the effect of Verb Type, and the effect of V erb Number subject noun. In fact, the materials of this experiment were the complete sentence versions of the target preambles used in Experiment 2, and so this experiment also serves to confirm the acceptability of those preambles in Experiment 2. Thus, in this experiment, participants read two lists of complete sentences and judged whether each sentence was acceptable in Persian or not. Also, they provided a

PAGE 116

116 response about whether they judged the subject of the sen tence as a "do er", something that was "acted on", or neither of these. E ffect of v erb t ype The results of this study showed that Verb Type had a significant effect on th e acceptability of mismatched verbs. The participants accepted more than 90% of the mismatched verbs in groups (A, B, F, G, and H) with inanimate plural subject nouns but they accepted 76% of the target sentences with verbs in group judgments for the ac ceptability of the target sentences with the Verb Types (A, B, F, G, and H) were almost the same but they were different for the Verb Types (C, D, and E). Also, the verb types (C, D, and E) were different from each other Considering the thematic role that different verb type s assign to their subject nouns, we may conclude that the Verb Type and the thematic role that the verb assigns to its subject have an important role in the optional subject verb agreement. This result is consistent with the result of t he effect of Verb type in Role interpretation which showed the significant effect of verb type. More specifically the verb type showed that the participants perceived almost the same porti on of the acted on thematic role for the subjects of verb groups (F & H). The verb groups (C, D) are similar and show very little proportion of acted on role. Group (D and E) show almost no acted on proportion and are similar. But the effect of verb in thi s goal is somehow different from the acceptability part as the proportion of acted on for the verb type (group A) is different from verb types (F & H), while it is considerably different from verb type (group B) and all the other verb

PAGE 117

117 types (the verb grou ps A, B, F, H showed similar range of unacceptability of mismatched verb). Considering the similarities and differences of the result of rate of unacceptability and proportions of acted on role for the subject noun, the verb groups may be classified in th C & D with agent like subjects, (3) verb type E with agent subjects, and (4) verb types F & H with patient subject. Effect of v erb t ense Neither Verb Tense nor the interaction of Verb Tense and Verb Type had any effective role in the acceptability of the mismatched verbs. This shows that the participants did not show any difference in accepting the past and present tense target sentences with inanimate plural subject noun and mismatched form of the verb ( singular). Verb tense did not show any effect in the interpretation of the thematic role of the subject noun either. Effect o f v erb n umber acceptance of the sentences; the participants accepted all the sentences with the matched (plural verbs) but they accepted different proportions of the sentences with mismatched verbs depending on the types of the verb. Verb Number also showed a significant effect on the pa nouns. They perceived more acted on role for the inanimate plural subject nouns when the verbs were singular rather than plural. Crucially, the effect of Verb Type and the thematic role it assi gns to its subject noun appear to be related to the semantic meaning rather than grammatical meaning. This implies the effect of semantic factor; the factor other than purely syntactic role. This

PAGE 118

118 fact that the Verb Type and the thematic role it gives to th e subject noun could influence agreement is most consistent with a Maximalist view and constraint based account in which additional, non syntactic factors may influence agreement. Conclusion As predicted, the results suggest the effect of Verb Type in the acceptability of sentences with mismatched verbs. Also, the Verb Type was a significant factor in interpreting the thematic role of the subject nouns. The analysis did not show any effect of Verb Tense nor the interaction of Verb Type and Verb Tense in re jecting the sentences with mismatched verbs, nor did the verb tense showed to be a significant factor in interpretation of the role of the subject nouns. Both acceptability and role interpretation showed the effect of Verb Number. The participants accept ed all the sentences with matched verbs but accepted different proportions of mismatched verbs for verbs of different group types. For role Interpretation, the participants interpreted more acted on roles for subjects of the sentences with mismatched verbs rather than matched ones. This finding is very interesting as it shows that the participants are unknowingly more likely to interpret patient thematic roles for subject nouns if the verb of the sentence is singular. The effect of verb type in rejecting th e sentences with mismatched form of the verb and interpreting the thematic role the verb assigns to its subject are semantic factors and consistent with the maximalist, constraint based account which argues that in addition to syntactic information other f actors including semantic factors influence agreement during grammatical encoding and that the agreement is not governed solely by syntactic information.

PAGE 119

119 The results of Role interpretation part showed the significant effect of verb type; as the participan ts interpret different portions of acted on roles for the subject nous with classified into verb types which give patient, patient like, agent, and agent like roles to their subject nouns.

PAGE 120

120 Table 3 1 Acceptability, test of within subject effects (per participants) Source df Mean Square F Sig. V erb type 7 8.300 353.047 .000 Error (verbtype) 189 .024 T ense 1 .045 2.420 .131 Error (tense) 27 .019 verbtype tense 7 .020 1.217 .295 Error (verbtype*tense) 189 .017

PAGE 121

121 Table 3 2 Acceptability, t est of within subject effects (per items) Source D f Mean Square F Sig. Tense 1 .018 1.269 .263 tense verb type 7 .007 .499 .833 Error(tense) 104 .014

PAGE 122

122 Table 3 3 Mean proportion of unacceptability of each Verb Type Verb Type Mean Std. Error A .039 .014 B .071 .020 C .244 .041 D .801 .035 E .964 .011 F .016 .009 G .027 .013 H .031 .017

PAGE 123

123 Table 3 4 Pairwise comparison of the mean proportion of unacceptability of mismatched verb for each Verb Type (I) verb type (J) verb type Mean Difference (I J) Std. Error A B .033 .020 C .205 .041 D .762 .039 E .926 .016 F .022 .013 G .012 .016 H .007 .018 B A .033 .020 C .173 .042 D .729 .033 E .893 .019 F .055 .020 G .045 .019 H .040 .021 C A .205 .041 B .173 .042 D .557 .047 E .720 .043 F .228 .040 G .217 .040 H .213 .036 D A .762 .039 B .729 .033 C .557 .047 E .164 .035 F .784 .034 G .774 .032 H .769 .033 E A .926 .016 B .893 .019 C .720 .043 D .164 .035 F .948 .013 G .938 .018 H .933 .018 Based on estimated marginal means a. Adjustment for multiple comparisons: Benferroni. *. The mean differe nce is significant at the .05 level

PAGE 124

124 Table 3 4 Continued (I) verb type (J) verb type Mean Difference (I J) Std. Error F A .022 .013 B .055 .020 C .228 .040 D .784 .034 E .948 .013 G .010 .013 H .015 .013 G A .012 .016 B .04 5 .019 C .217 .040 D .774 .032 E .938 .018 F .010 .013 H .004 .016 H A .007 .018 B .040 .021 C .213 .036 D .769 .033 E .933 .018 F .015 .013 G .004 .016 Based on estimated marginal means a. Adjustment for multiple c omparisons: Benferroni. *. The mean difference is significant at the .05 level

PAGE 125

125 Table 3 5 Role interpretation, t est of within subject effects (per participant) Source df Mean Square F Sig. V erb type 6 18.016 549.120 .000 Error(verb type) 126 .033 Verb T ense 1 .063 2.839 .107 Error(tense) 21 .022 N umber 1 .124 4.745 .041 Error(number) 21 .026 V erb type verb tense 6 .017 .682 .664 Error(verb type verb tense) 126 .025 V erb type verb number 6 .009 .28 9 .941 Error(verb type verb n umber) 126 .032 Verb tense verb number 1 .000 .004 .951 Error( verb tense verb number) 21 .027 verb type verb tense verb number 6 .015 .840 .541 Error(verb type* verb tense verb number) 126 .01 8

PAGE 126

126 Table 3 6 Role interpretation, t est of within subject effects (per items) Source D f Mean Square F Sig. Verb tense 1 .078 6.535 .012 Verb tense verb type 6 .027 2.257 .045 Error(v erb tense) 90 .012 Verb n umber 1 .066 4.523 .036 Verb number verb type 6 .008 .533 .782 Error(v erb number ) 90 .015 Verb tense v erb number 1 .000 .028 .868 Verb tense v erb number verb type 6 .013 .954 .461 Error(v erb tense v erb number ) 90 .014

PAGE 127

127 Table 3 7 Mean proportion of patient role interpreta tion for different Verb Type Verb Type Mean Std. Error A 0.686 .038 B 0.184 .033 C 0.046 .013 D 0.012 .007 E 0.000 .000 F 0.988 .007 G ----H 0.966 .013

PAGE 128

128 Table 3 8 Pairwise compa rison of the mean proportion of acted on interpretation of t he subject nouns for each Verb Type (I) verb type (J) verb type Mean Difference (I J) Std. Error Sig. a A B .502 .042 .000 C .639 .038 .000 D .673 .036 .000 E .686 .038 .000 F .302 .038 .000 H .280 .034 .000 B A .502 .042 .000 C .137 .034 .012 D .171 .033 .001 E .184 .033 .000 F .804 .034 .000 H .782 .031 .000 C A .639 .038 .000 B .137 .034 .012 D .034 .016 .847 E .046 .013 .048 F .941 .013 .000 H .920 .017 .000 D A .673 .036 .000 B .1 71 .033 .001 C .034 .016 .847 E .012 .007 1.000 F .975 .012 .000 H .954 .016 .000 E A .686 .038 .000 B .184 .033 .000 C .046 .013 .048 D .012 .007 1.000 F .988 .007 .000 H .966 .013 .000

PAGE 129

129 Table 3 8 Continued (I) v erb type (J) verb type Mean Difference (I J) Std. Error Sig. a F A .302 .038 .000 B .804 .034 .000 C .941 .013 .000 D .975 .012 .000 E .988 .007 .000 H .022 .013 1.000 H A .280 .034 .000 B .782 .031 .000 C .920 .017 .000 D .954 .016 .000 E .966 .013 .000 F .022 .013 1.000

PAGE 130

130 Table 3 9 Mean proportion of patient role interpretation for different Verb Tense T ense Mean Std. Error Present .402 .012 P ast .422 .011

PAGE 131

131 Table 3 10 P airwise comparison of patient interpreta tion for different Verb Tense (I) tense (J) tense Mean Difference (I J) Std. Error Sig. a present past .020 .012 .107 P ast present .020 .012 .107

PAGE 132

132 Table 3 11 Mean proportion of patient interpretation for different verb number N umber Mean Std. E rror P lural .397 .013 S ingular .426 .010

PAGE 133

133 Table 3 12 P airwise comparison of patient interpretation for different verb number (I) number (J) number Mean Difference (I J) Std. Error Sig. a Plural Singulat .028 .013 .041 singular Plura l .028 .013 .041

PAGE 134

134 Table 3 13 P atient interpretation for subjects ( interaction of verb tense and verb number ) verb type number Mean Std. Error A PL .657 .051 SG .714 .054 B PL .163 .043 SG .205 .040 C PL .036 .017 SG .057 .019 D PL .00 8 .008 SG .017 .009 E PL .000 .000 SG .000 .000 F PL .975 .014 SG 1.000 .000 H PL .943 .025 SG .989 .011

PAGE 135

135 CHAPTER 4 EXPERIMENT 2 E FFECT OF VERB TYPE AND TENSE Overview As discussed in Chapters 1 and 3, some Persian grammarians and linguists have suggested that different kinds of verbs may have an impact on optional subject verb number agreement in Persian (Lotfi, 2006; Saasat, 1996). Also, Lotfi (2006) argued that Persian speakers prefer to use plural verbs when the tense of the sentence is f uture and not past. The goal of this experiment was thus to investigate the effect of verb type and verb tense in the production of optional subject verb agreement in Persian. More specifically, this experiment was designed to test whether the tense of th e sentence (present vs. past) causes speakers to produce a higher proportion of mismatched (i.e. singular) verbs or not. In the same way, the effect of verb type was tested to investigate whe ther participants produce a higher proportion of mismatched verbs with certain verb types compared to others. It is worth repeating from Chapter 3 that the materials of this experiment were tested in the judgment task reported in Chapter 3, in which the participants accepted all the target sentences with plural verbs ei ther past or present and belonging to any verb group. But, with mismatched verbs, participants accepted different proportions of sentences depending on the verb type, with verbs interpreted as assigning more patient like roles to their subjects have the hi ghest rates of acceptability. Participants did not show any acceptability different for past vs. present tense. Now, in this completion task, participants were provided with two series of preambles (incomplete versions of the target sentences in Chapter 3) and they were

PAGE 136

136 instructed to repeat each preamble and then complete the sentence with their own verbs in present or past tense. Method Participants The participants were forty six Iranian native speakers of Persian (29 males and 17 females) between 22 4 3 years old (mean 28 years). They were members of the University of Florida community, and participated in the study voluntarily. The participants were late learners of English who had learned this language for the purpose of education and did not have sig nificant knowledge of any other language. All participants had normal or corrected vision and did not have any cognitive or neurological impairment. The participants of this experiment also participated in Experiment 5 (object attraction, cf. Chapter 7), w hich was conducted at the same time. Materials The materials of this experiment were almost the same as the materials of Experiment 1 (Chapter 3). There was only one difference: the sentences in Experiment 1 were complete sentences (consisting of inanimate plural subject noun + nonverbal + verbal part of the combined compound verbs) while the materials of this experiment were preambles consisting of [inanimate plural subject noun + nonverbal part of the compound verbs]. The reason for the difference is that in the judgment task the participants were supposed to read the sentences and judge whether they were acceptable sentences, but in this completion task they were supposed to repeat the preamble and then complete the sentences by adding the verbal part of the compound verb.

PAGE 137

137 The materials for this experiment were thus one hundred and twelve target preambles. The preambles were classified into the same eight groups as in the judgment task: (A) unaccusative verbs, (B) Entity specific inchoative verbs, (C) ver bs of emission, (D) verbs of instrument, (E) verbs of agency, (F) passive verbs, (G) stative coded in order to test the effect of the verb type in the production of m ismatched form (singular) verb. There were 14 target preambles in each verb group. Table 4 1 shows a sample preamble of each verb type along with the possible responses in present/past tense with singular/plural verbs. For the English equivalent (fourth li ne), only past plural is provided. The distracter materials were also the same as those in Experiment 1. There was complete sentences, while the distracter materials in th is experiment were preambles with similar structure as the critical items. The target sentences were divided into two compatible lists (one list of preambles to be completed with past tense verbs and the other with present tense). In each list, there were fifty six target preambles (seven of each condition) and forty filler preambles. Participants saw both of the lists and completed the sentences of one list with present tense and the other list with past tense verbs (in this way the participants read and complete all the sentences). The order of the trials was arranged pseudo randomly so that there were no adjacent target sentences within the same condition. Each list was read by twenty two participants.

PAGE 138

138 A list of the target and filler items in gloss and in Persian are available in Appendix B. Because understanding the literal meaning (the third line of the gloss) could be difficult for non native speakers of Persian, an English grammatical equivalent of the complete sentence was added in the forth line. Procedure Participants were tested individually and gave informed consent before participating. They sat in front of an Apple Mac computer equipped with a microphone running Psyscope X. The experiment started with instructions and four practice trials. Whe n participants indicated they were ready, the experiment began. As mentioned above, in order to test the effect of Verb Tense, the participants were instructed to imagine that the events of the preambles in one of the lists had happened the day before but for the other list the events happened on the same day. In this way, they automatically only used past tense verbs in one list and present tense verbs in another list. List A and B were used alternatively for present or past tense so half of the participa nts repeated and completed the preambles of list A with past tense and list B with present tense and the other half did opposite. Also, in order to avoid an effect of order with the tenses, half of the participants did present tense first, then past. The o ther half of the participants did the opposite order. During a break between the two lists, all the participants participated in a different completion task (Experiment 5). Even with the additional task in between, participants had a few minutes break betw een each list. In each trial, the preamble appeared in black text at the center of the computer screen on a white background. The participants were instructed to repeat the preamble and then complete the sentence aloud. Then, they pushed any key to conti nue, which

PAGE 139

139 caused a signal (++++++) to appear on the screen. The participants pressed a key again when they were ready to proceed to the next trial. After pressing the key, it would take one second until the next trial would appear. This would give enough time for me responses in any unclear cases. Following the approved IRB protocol, particip voice files were deleted from the computer after the investigator recorded the Design and Data Analysis The verbs produced by the participants were entered into an Excel spreadsheet, coded one or zero based on the singularity (1) or plurality (0) of the produced verbs, resulting in a proportion of singular verb (subject mismatching) responses. Recall that the question for this experiment is whether there are effects of Verb Type and Verb Tense in the production of subject verb agreement. The dependent variable is the proportion of singular verbs out of all items. As already noted above, there were two independent, within subject factors: Verb Type (with eight levels, (A) unaccusative verbs, (B) Entity specific i nchoative verbs, (C) verbs of emission, (D) vs. Past). Sets of by participants (F1) and by items (F2) Repeated Measures ANOVAs with the eight level factor Verb Type and two level factor Verb Tense were performed. In addition, planned paired sample t tests were performed. An alpha level of .05 was used for all statistical tests. One

PAGE 140

140 participant had difficulty performing the task in a timely fashion (she had been living in another state for two years before moving to Florida and had not had access to Iranian communities to use P ersian). The data of another participant was also excluded because the participant was using formal language rather than informal spoken Persian Predictions Regarding the effect of tense, it w as predicted that more mismatched (i.e. singular) verbs would be produced in sentences with past tense compared to sentences with present tense. For the effect of verb type, participants were predicted to produce more mismatched verbs in A, B, F, G and H (with patient/ patient like/ or no role subject) than in C (subject of emission verbs), D (intermediary instrument subject), or E (agent subject). Results The analysis revealed a main effect of Verb Type (F1(7, 301)=55,336, MSE=0.032, P<0.01), a main effec t of Verb Tense (F1(1, 43)=47,127, MSE=0.100, P<0.01; F2(1, 104)=207,513, MSE=0.007, P<0.01), and an interaction between Verb Type and Verb Tense (F1(7, 301)=9,018, MSE=0.025, P<0.01; F2(7, 104)=10,081, MSE=0.007, P<0.01; ). The results are shown in Tables 4 2 and 4 3. Table 4 mismatched (singular) form of the verb in present and past tense sentences. As Table 4 4 indicates, 17% of the produced sentences in the present tense and 33% of those in the past tense had mismatched verbs. This means that the participants

PAGE 141

141 produced nearly double the number of mismatched verbs in the past tense sentences compared to the present tense sentences. matched (singular) form of the verb for the verbs of each group (A, B, C, D, E, F, G, and H) are given in Table 4 5. groups (A) unaccusative verbs, (B) Entity specific inchoative verbs, (F) passive verbs, 0.38). This means that the participants produced mismatched verbs in 32% to 38% of sentences in those verb groups. Considering the thematic role of the subject nouns in these types of the verbs, (patient, patient like, or no role for stative verbs) we may conclude that Verb Type and the thematic role that the verb assigns to its subject have a significant role in optional subject verb agreeme nt. However, the mean proportion verbs is different for the other three verb groups: (C) verbs of emission (with 0.17), (D) verbs of instrument (with 0.06), and (E) verbs of agency (with 0.02). Considering the thematic role that the verbs in these groups assign to their subject nouns, we may conclude that the Verb Type and the thematic role that the verb assigns to its subject has an effect in optional subject verb agreement. (The verbs of agency (E) assign in the previous chapter (Chapter 3), it is difficult to say what thematic role the verbs of emission (C) assign to their subject nouns because the subjects are inanimate n ouns and do not have voluntary control on the action.)

PAGE 142

142 agency mismatched verbs for the animate ve rbs (e.g. cry, dance, walk). This 0.02 production of the mismatched form of the verb is because of the stereotypical meaning of one of the target items, which was explained in Chapter 3. Table 4 smatched verbs, with the interaction of Verb Tense and Verb Type. The proportion of the mismatched verbs in each group was different for past vs. present tense. Generally, the proportion of singular verbs produced with the past tense is about twice that of the present tense. This interaction of verb type and tense was most clear in verb groups A (unaccusative verbs, 2.18), B (entity specific inchoative verbs, 2.07), D (verbs of the biggest effect. The interaction of verb type and tense was smaller in verb groups C (verbs of emission, 1.59) E (verbs of agency, 1.77) F (passive verbs, 1.88) and H (inchoative t. This suggests that the effect of tense is stronger in some verb groups such as stative auxiliary (groups G) and less effective in verbs of emission (group C), but it is not entirely clear why. Discussion In this study, the effects of Verb Type and Verb Tense in the production of optional subject verb agreement were tested. The optionality of subject verb number agreement of Persian (for inanimate plural subjects) and the use of compound verbs created exceptional conditions for testing the effect of Verb Type and the thematic role it assigns to the subject. All experimental materials were grammatical sentences and

PAGE 143

143 there was no need to use elicitation of errors or the ambiguity of collective or distributive nouns. Further, the structure of compound verbs (consist of nonverbal + verbal part) in the target preambles provided the opportunity to give the nonverbal part of the verb in the preamble and ask the participants to complete the sentence with a verb. In this way, participants had no choice other than to complete sentences with the verbal part which was predicted. It is worth noting that the type of the verb and the thematic role it assigns to the subject noun is determined by the verbal part of the compound verb. The results of this study show th at Verb Tense had an important impact on the production of optional subject verb agreement. The participants produced mismatched (singular) verbs in the past tense almost twice as much as they produced them in the present tense, as was predicted by Lotfi ( 2006). The experiments of this dissertation did not investigate the reason behind the effect of verb tense. Lotfi (2006: p137) has predicted that Persian speakers use more plural verbs in future sentences. He believes pparently, autonomy in action i s more viable when the course of events has not Also, we saw a significant effect of Verb Type in the production of optional agreement, in which participants produced between 32% to 38% mismatched verbs in preambles with inanimate plu ral subject nouns and verb groups (A, B, F, G, and H) but produced only 17 % mismatched verbs with verbs in preambles with group (C), 6% with group (D), and only 2% with group (E) (Table 4). This comparison of the mean proportion of the mismatched verbs sho ws that the participants produced almost the same proportion of mismatched verbs with the Verb Types (A, B, F, G, and H) but different proportions with the Verb Types (C, D, and E). Considering the thematic role

PAGE 144

144 that different verb types assign to their su bject nouns, we may conclude that the Verb Type and the thematic role the verb assigns to its subject influences the choice of verb number in optional agreement. Interestingly, there was also an interaction of Verb Tense and Verb Type, showing that verbs w ere impacted to differing degrees by tense. Crucially, the effect of Verb Type and the thematic role it assigns to its subject noun appear to be related to semantic meaning rather than grammatical meaning. This results, showing the impact of a semantic fa ctor, is most consistent with a Maximalist view and constraint based account in which additional, non syntactic factors may influence agreement. Conclusion As predicted, the results of the study showed that Verb Tense was a significant factor in the prod uction of optional subject verb agreement; the participants produced more mismatched (singular) verbs in the past tense sentences comparing to the present tense sentences, as was predicted by Lotfi (2006). Also, the Verb Type and the thematic role the verb assigns to its subject was an influential factor in the production of optional subject verb agreement. Participants produced almo st the same portion of mismatched (singular) verb with the verbs in groups (A, B, F, G, and H) in which the verb assigns patie nt/patient like/ or no role to the subject. But they produced different portions of the mismatched verb with the other verb groups. In these three later verb groups, the verbs assign different thematic roles other se results suggest the effect of Verb Type and the thematic role they assign to their subjects in the production of optional subject verb agreement.

PAGE 145

145 These results, that production of a mismatched form of the verb in optional subject verb number agreement is affected by the Verb Tense and Verb Type and the thematic role the verb assigns to its subject are semantic factors and consistent with the maximalist view and constraint based account, which argues that in addition to syntactic information other factor s, including semantic factors, influence agreement during grammatical encoding and that the agreement is not governed solely by syntactic information.

PAGE 146

146 Table 4 1 S ample preamble of v erb t ype v erb t ense experiment & possible response Verb Type Gloss poss ible target response A tnab ha mi xor e mi xor n xord xord n rope pl IND collide 3SG IND collide 3PL collide.Past.3SG collide.Past 3PL the ropes knot ... the ropes became knotted. B Lubia ha mi zn e mi zn n zd zd n bean pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the beans sprouting ... the beans sprouted. C pnjre ha mi kon e mi kon n krd krd n window pl IND DO 3S G IND DO 3PL do.Past.3SG do.Past 3PL the windows rattling ... the windows rattled. D bolndgoo ha mi kon e mi kon n krd krd n loudspeaker pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the loudspeaker s announcing ... the loudspeakers announced. E tr ha mi kon e mi kon n krd krd n perfume pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the perfumes crying ... the perfumes cried. F go labi ha mi sh e mi sh n shod shod n pear pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the pears were eaten.

PAGE 147

147 Table 4 1. Continued Verb Type Gloss possible target response G miz ha ye y n bud bud n table pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the tables ... wooden the tables were wooden. H rng ha mi sh e mi sh n shod shod n paint pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the paint ... dry the paint dried.

PAGE 148

148 Table 4 2 Test of within subject effect (per participants) Source D f Mean Square F Sig. t ense 1 4.708 47.127 .000 Error (tense) 43 .100 V erb T ype 7 1.798 55.336 .000 Error (verb type ) 301 .032 tense verb type 7 .229 9.018 .000 Error (tense* verb type ) 301 .025

PAGE 149

149 Table 4 3 Test of within subject effect (per items) Source Df Mean Square F Sig. Tense 1 1.499 207 .513 .000 tense VT 7 .073 10.081 .000 Error(tense) 104 .007

PAGE 150

150 Table 4 4 Mean proportion of SG verbs produced in present and past sentences Tense Mean Std. Error Present .165 .020 Past .328 .029

PAGE 151

151 Table 4 5 M ean proportion of SG verbs p roduced in different verb groups VT Mean Std. Error A .320 .029 B .328 .033 C .172 .022 D .060 .014 E .018 .005 F .383 .037 G .339 .036 H .352 .032

PAGE 152

152 Table 4 6 E ffect of interaction of Verb Type and Verb Tense in the production of mismatched ve rb s Verb Tense Verb Type Mean Std. Error Present A .201 .033 B .214 .033 C .133 .025 D .039 .017 E .013 .008 F .266 .040 G .208 .038 H .244 .033 Past A .438 .042 B .442 .043 C .211 .031 D .081 .019 E .023 .008 F .500 .049 G .471 .047 H .461 .044

PAGE 153

153 CHAPTER 5 EXPERIMENT 3 EFFECT OF UNITY OF SUBJECT NOUN Overview The goal of this experiment was to investigate the effect of unified vs. individualized conceptualization of the entities of subject nouns in optional subject verb agreement. Meshkat al dini (1987) and Lotfi (2006) proposed that the conceptualization of the entities of subject nouns, as either individual entities or as a unit, influences optional subject verb agreement in Persian and suggested further st udy to test the effect of this factor on the selection of singular vs. plural form of the verb for inanimate plural subject nouns. Based on their suggestions and an informal pilot study, it seems likely that Persian speakers do produce more singular verbs with inanimate plural subject head nouns when the entities of the subject head nouns are considered as a unit comparing to the situation when the entities of the subject head nouns are considered as individuals. In the examples below, Persian speakers see m to produce more singular verbs in cases where they consider the entities of the subject plural noun as a whole (e.g. a collection of leaves), while they prefer to use a plural verb in cases where they consider the entities of the subject nouns as individ ualized (e.g. each leaf on its own). (5.1) b rg ha xoshk shod /shod n leaf PL dry become.3SG /become 3PL This experiment was designed to test this hypothesis, asking if the Unity vs. Individuality of inanimate plural subject nou ns would have any effect in subject verb agreement. And, more specifically, whether the Unity vs. Individuality of inanimate plural

PAGE 154

154 subject nouns causes Persian native speakers to use more singular or plural forms of the verbs. There is some relevant data from English. As discussed in Chapter 2, Humphreys and Bock (2005) tested the effect of notional number in the subject verb agreement. In a completion task, they used the ambiguous nature of collective nouns to assess the impact of distributivity in minima lly contrastive contexts. Based on Gestalt more significant. They used different prepositions with a modifier to manipulate the notional number of for the ambiguous subject ( the gang gives the notion of significance. Using the gang on motorcycles the gang near motorcycles (5.2) a. The gang on b. The gang near Similar to previous completion task studies o n agreement, the participants were provided with a series of preambles and instructed to repeat each preamble and then complete the sentence in their own words. The results showed that significantly more plural verbs were used with the distributive concept ualization of the subject than the collective conceptualization. spatial separation vs. to investigate the effects of collective and distributive conceptualization of the subject in optional sub ject verb number agreement in Persian. Similar to Humphreys and Bock

PAGE 155

155 (2005), I used different prepositions with the modifiers of the inanimate plural subjects to Individuality vs. of the subject entities. However, because subject verb number agreement (for inanimate plural subjects) is optional in Persian and both singular and plural forms of the verb are grammatical, I did not need to rely on the ambiguity of collective nouns or the elicitation of errors. I used inanimate plural subje cts in normal grammatical sentences. Based on Meshkat al Bock (2005) and my own informal pilot study, an effect of subject Unity vs. Individuality in subject verb agreement was predicted. The partic ipants were expected to produce more mismatched verbs (i.e. singular verbs) with the inanimate plural subjects when they focus on the Unity of the elements of the subject nouns but to produce fewer mismatched verbs when they focus on the Individuality of t he elements. Method Participants The participants were twenty five Iranian native speakers of Persian (16 males and 9 females) between 22 30 years old (mean 26 years). They were members of University of Florida community, and participated in the study vo luntarily. The participants were late learners of English who had learned this language for the purpose of education and did not have significant knowledge of any other language. All participants had normal or corrected vision and did not have any cognitiv e or neurological impairment. The participants of this experiment also participated in Experiments 1 and 4 (testing the effect of subject concreteness). Experiment 1 was conducted more than one year before this experiment and Experiment 4 was conducted at the same time.

PAGE 156

156 Materials Twenty four sets of target and seventy two distracter preambles were constructed. Twenty four items out of seventy two distracter items were the target preambles of another experiment (the effect of Concreteness vs. Abstractness o f the subject head nouns on optional agreement see Chapter 6). The target preambles consisted of an inanimate plural subject noun, a post modifying prepositional phrase and either a past participle or an adjective, creating two similar structures, passive and inchoative auxiliary (see vs. highlight the collective vs. distributive conceptualizations of the subjects. The prepositions were manipulated so that the prepositional phrases modifying the subject head noun emphasized the Unity (e.g. near collective group ) or Individuality (e.g. on ch each cup is seen as a separate entity ) of the entities in subject positions. The thematic role of the subject nouns was patient. This means that the subjects were acted on inanimate nouns although they were the subject of the sentence in the absence of an agent noun. As discussed above, the participants were expected to produce more plural verbs with the distributive conceptualization of the subject. But, this result could be due to an effect of attraction (e.g. the plurality of the local modifier noun) So, a singular modifier preceded by a spatial collection preposition was also examined in a compatible preamble (e.g. near ) Thus, manipulating the Preposition and Number of the local noun created three kinds of subject phrases: Ind ividualized subject with plural modifier (IP), Unified subject with plural modifier (UP), and Unified subject

PAGE 157

157 with singular modifier (US). Table 5 1 shows an example of the conditions of subject nouns and the related prepositional phrases of the preambles. Similar to the previous experiments in this dissertation, the compound verb structures in Persian enabled me to provide the participants with the non verbal part of the verbs and restrict the participants to complete the sentences with the auxiliary verb (/ shod/ became.3SG or / shod n/ became 3PL ). Although the participants were told to complete the sentences with any verb they liked, the preambles and the instructions were designed in a way to restrict them to use only the auxiliary verb ( shod, became.3SG or shod n, become 3PL ). This allowed us to avoid the any effects of different verbs or verb types. The auxiliary / shod is the auxiliary verb used in passive and inchoative auxiliary structures So, the preambles in the passive structure co nsisted of subject noun+ modifying prepositional phrase +past participle while the inchoative auxiliary structure consisted of subject noun + modifying prepositional phrase + adjective. Both types of preambles could be completed by the auxiliary verb / shod Preparing sets of three preambles with inanimate plural subject and modifying prepositional phrase in equal conditions except for the Type of the pr eposition and Number of the local noun (e.g., the gems on the rings the gems near the rin gs, and the gems near the ring) led me to use passive sentences. But passive sentences in English are either clearly passive or inchoative auxiliary 1 or ambiguous between these two in Persian ( see Chapter 1 ). 1 By inchoative auxiliary I mean the inchoative compound verbs in Persian that consist of non verbal part plus auxiliary shod have different verbal pa rts, i. e. the auxiliary shod the verbal part of all the compound verbs.

PAGE 158

15 8 Although the results of the Verb Type experim ent (Chapter 4) showed the similar effects of passive and inchoative auxiliary verbs on the subject verb agreement in Persian (the participants produced similar percentage of SG verbs with the two structures ) and they are equivalent of passive structure in English, the preambles of each structure were coded differently (to compare their effects in different structures/ in different experiment). The code (V) was used for passive preambles and (J) for inchoative auxiliary ones. So there were two factors: (1) Type of the Preposition phrase (individual plural, unit plural, and unit singular) and (2) the Structure (passive vs. inchoative auxiliary ). This resulted in 6 conditions for the target sentences (IPV, IPj, UPV, UPJ, USV, USJ). A set of sample target prea mbles for one item and possible target responses of each condition is shown in Table 5 2. The participants were instructed to read the preambles out loud and complete each sentence with their own auxiliary verb, using informal, spoken Persian. Thus, the pa rticipants just needed to complete the sentence with a specific auxiliary verb / shodn/ In order to restrict the participants to use only the verb /shodn/ to get the desired structure, the participants were requested not to use th e verb which could also be used in auxiliary structure ). It is worth mentioning that the effect of the stative auxiliary structure (subject noun + adjective + the verb ) on optional subject verb agreement was similar to the effect of passive and inchoative auxiliary structure. Also, in order to keep the effect of tense constant, the participants were told to imagine that all the events of the sentences happened the day before, which allowed them to naturally use the past te nse verbs.

PAGE 159

159 The filler items included different kinds of structures. They consisted of a singular or plural (in)animate subject noun, optional singular/plural (in)animate direct object noun, the object marker /ra/ (when there was an object), optional indir ect object, and the nonverbal component of a (in)transitive compound verb. Using a Latin square design, the target sentences of each of the conditions (passive or inchoative auxiliary ) were divided into three compatible experimental lists. In each list, there were eight preambles of each condition (with IP, UP, and US subject prepositional phrase). Four of the preambles were with copular structure and the other four preambles with passive. No item repeated so that for each list, each participant saw every item exactly once, and across all the lists each item appeared once in each condition. The forty eight filler preambles were the same in all the three lists. The order of the trials was arranged pseudo randomly so that target sentences from the same cond ition did not appear next to the each other in the list. Each list was read by eight participants. List of the target and filler items in gloss and in Persian language along with the related condition are available in Appendices C at the end of the dissert ation. Since understanding the literal meaning (the third line of the gloss) could be difficult for non native speakers of Persian, an English grammatical equivale nt of the complete sentence was added in the forth line. Procedure Participants were tested individually. After giving informed consent, participants sat in front of an Apple MacBook Pro computer equipped with a microphone and running Psyscope X. The experiment started with a set of instruction and four practice trials. As mentioned above, in ord er to avoid the effect of tense, the participants were

PAGE 160

160 told to imagine that all the events of the sentences happened the day before. When participants indicated that they were ready, the experiment began. In each trial, the preamble appeared in black text at the center of the computer screen on a white background. The participants were instructed to repeat the preamble, complete the sentence aloud, and then should push a button to continue. At this point a signal (++++++) appeared on the screen. The parti cipants were supposed to push a button again when they were ready to proceed to the next trial. After pushing the button, it would take one second until the next trial would appear. This would give enough time answers in a related checklist. Throughout ce files were deleted from the computer when the Design and Data Analysis The verbs produced by the participants were coded one or zero based on the singularity (1) or plurality (0) forms o f the verbs respectively. Recall that the question for this experiment is whether there is an effect of the Unity vs. Individuality (of the subject phrase) in the production of subject verb agreement. The dependent variable is thus the proportion of singul ar verbs out of all items. As already noted above, there were two independent, within subject factors: Structure (with two levels, passive and inchoative auxiliary ) and Type of the Prepositional Phrase (with three levels, Plural prepositional phrase empha sizing the Individuality of subject head noun (IP), Plural prepositional phrase emphasizing the Unity of the subject head noun (UP), Singular prepositional phrase emphasizing the Unity of the subject head noun (US)).

PAGE 161

161 Sets of by participants (F1) and by ite ms (F2) Repeated Measures ANOVAs with the two level factor Types of Structure and three level factor Types of the Prepositional phrase performed. In addition, planned paired sample t tests were performed. An alpha level of .05 was used for all statistical analysis because the participant had difficulty performing the task in informal Persian Results There were two independent, within subj ect factors: Structure (with two levels, passive and inchoative auxiliary ) and Type of the Prepositional Phrase (with three levels, Individualized Plural, Unified Plural, and Unified Singular). IPV: UPV: ems near the rings USV: IPV: UPV: USV: This analysis revealed a main effect of Types of Prepositional phrase ( F 1 (2, 46) = 4.449, MSE = 0.107, p < 0.05; F 2 (2, 44) = 8.338, MSE = 0.028, p < 0.05), with singular or plural prepositional phrase emphasizing on the unity of the subject entities. There was no effect of Struct ure (F1 = 2.67) or an interaction between Types of Prepositional phrase and Structure ( F1 <1, F2 < 1). This shows that the structure of the sentence did not have an effect on the singularity or plurality form of the verbs that the participants produced. Th ese results are reflected in Table 5 3 and 5 4

PAGE 162

162 These results indicate that the conceptualization (Unity vs. Individuality) of the enti ties of the subject nouns was an effective factor in the optional subject verb agreement in Persian. This means that under nearly equal conditions with inanimate plural subjects, the participants produced more singular verbs when the preposition of of the entities of the subjects. On the o ther hand, the lack of a significant effect of Structure Type suggests that the passive and inchoative auxiliary structures do not have different effects in optional subject verb agreement (similar to the result of Experiment 2, Chapter 4). Hence, the part icipants produced more mismatched (singular) verbs for the Unified inanimate subject nouns (UPJ, UPV, USJ, and USV) compared to the Individualized inanimate plural subject nouns (IPj and IPV). The above result showed that the participants produced more si ngular verbs with unified conceptualization of the subject. In order to ensure that this result is due to the effect of Unity vs. individuality of entities of the subject noun and not due to the attraction effect of the local noun (the effect of the plural ity of the local modifier noun) on the verb, the proportions of singular verbs produced with Unified Plural and Unified Singular were compared. Table 5 5 shows t he proportion of singular verb usage with different prepositional phrase. According to this tab le, there is not much difference between the proportion of singular verbs produced with the Unified Plural (0.41) and Unified Singular (0.44) propositional phrases, compared to the proportion of singular verbs produced with Individualized Plural propositio nal phrases (0.26). These results show that the participants produced 0.44% singular verbs with modifying prepositional phrase that emphasized the unity of the elements of the plural

PAGE 163

163 subject head noun while the local noun of the prepositional phrase was singular (e.g. the gems near the ring ) Participants produced 0.41% singular verbs with prepositional phrases which emphasized the unity of the elements of the plural subject head nouns while the local noun of the prepositional phrase was plural (e.g. the gems near the ) And, finally, they produced 0.26% singular verbs with modifying prepositional phrases which emphasized the individuality of the elements of the plural subject head noun while the local noun of the prepositional phrase was plural (e. g. the gems on the rings ) So, comparing the proportion of singular verbs produced with a singular vs. plural modifier prece ded by a spatial collection preposition shows that there is an effect of attraction (the effect of plurality of the local modifier n oun) but that this effect is much less than the effect of Unity vs. Individuality of entities of subject nouns. Table 5 6 shows the results of a pairwise comparison of the different prepositional phrases. The difference between Individual Plural and Unity plural is significant while the difference between Unity plural and Unity singular is not. Table 5 7 shows the mean proportion of mismatched verb used with the three prepositional phrase and two structures. Discussion The results of this study showed that the participants produced more mismatched verbs (singular verbs) with inanimate plural subject nouns in the condition that the preposition of the prepositional phrase modifying the subject head noun was emphasizing on the Individuality of the entities of t he subject noun (e.g. the gems on the rings ) compared to the condition in which the preposition of the prepositional phrase modifying he subject head noun was emphasizing on the Unity of the entities of the subject noun (e.g. the gems near the rings ). This result indicates that the unity vs.

PAGE 164

164 Individuality conceptualization of the inanimate plural subject entities has a significant role in the production of singular/plural form of the related verb. In order to check that this effect is due to the unity vs. individuality conceptualization of the subject nouns and not due to the attraction effect of the plurali ty of the local noun of the prepositional phrase modifying the subject, the proportions of the singular verbs produced in sentences with singular vs. pl ural local nouns were compared. The result showed that the plurality of the local nouns had an attractio n effect on optional subject verb agreement (more singular verbs were produced with Unified Singular comparing to Unified Plural), but this effect was m uch less than the effect of Unity vs. Individuality of the subject. Finally, the results of this study did not show any effect of structure type. This means that passive and inchoative auxiliary structures had almost the same effect on form of the verb (to produce singular or plural verb). Crucially, the effect of Unity vs. Individuality of the entities of the subject appears to be related to the conceptual meaning rather than grammatical meaning. This implies the influence of a semantic factor, rather than a purely syntactic influence. In the target items of this experiment, the inanimate plural subject nouns are the same. What matters is that the entities of one subject are conceptualized as a unit while the entities of another subject are conceptualized a s individuals by the preposition of their modifying prepositional phrases. Hence, the participants prefer to produce more singular verbs with the inanimate plural subject nouns conceptualized as a unit rather than individual ones. The fact that conceptual meaning could influence agreement is most consistent

PAGE 165

165 with a Maximalist view and constraint account in which additional, non syntactic factors may influence agreement. Conclusion In this study, the effect of unified vs. individualized conceptualization of inanimate plural subject nouns in optional subject verb agreement in Persian was tested. The optional subject verb number agreement of Persian (for the inanimate plural subjects) creates an exceptional condition to test the effect of conceptual meaning of subject nouns in subject verb agreement in otherwise almost equal conditions, where all the materials were grammatical sentences and there was no need to use elicitation of errors neither ambiguity of distributive or collective nouns. Creating preambles w ith inanimate plural subject head noun, a modifying prepositional phrase, and nonverbal part of the compound verb led me to passive and inchoative auxiliary structures. I coded these two structures differently, resulting in two factors: Structure Type and Type of the Prepositional phrase. The prepositions emphasized either the Unity or Individuality of the entities of the subject. In order to test the attraction effect of the plurality of the local noun (the noun of the prepositional phrase), the number of the local noun were manipulated. So, there were three Types of Prepositional phrase; Individualized Plural, Unified Plural, and Unified Singular. The participants read the preambles and complete the sentences with the verbal part of the compounds. As predi cted, the results of the study showed that the singularity vs. plurality of the produced verbs was affec ted by the Types of the Prepositional phrase. More specifically, the participants produced more singular verbs with Unified prepositional phrases rathe r than Individualized ones. Comparing the proportion of singular produced

PAGE 166

166 verbs with Unified Singular and Unified Plural Prepositional phrases showed a small attraction effect of local nouns (the noun of the prepositional phrase) but this effect was much l ess than the effect of Unity vs. Individuality of the subject noun. This result, that the optional subject verb number agreement is affected by Unity vs. Individuality conceptualization of the entities of the subject noun, supports the Maximalist view and constraint account which argues that both conceptual and syntactic information influence agreement during grammatical encoding and the agreement is not governed solely by syntactic information.

PAGE 167

167 Table 5 1 C onditions of subject phrases C ondition examp le I ndividual plural negin ha ye ruy e ngoshtr ha gem PL EZ on EZ ring PL the gems on the rings U nity Plural negin ha ye nzdik e ngoshtr ha gem PL EZ near EZ ring PL the gems near the rings Unity Singular negin ha ye nzdik e ngoshtr gem PL EZ near EZ ring the gems near the ring

PAGE 168

168 Table 5 2 S ample preamble and possible target response for each condition condition Preamble possible target response IPV negin ha ye ruy e ngoshtr ha psndid e shod /shod n gem PL EZ on EZ ring PL chose pp became.3SG /became 3PL the gems on the rings were chosen. UPV negin ha ye nzdik e ngoshtr ha psndid e shod /shod n gem PL EZ near EZ ring PL chose pp became.3SG /became 3PL the gems near the rings were chosen. USV negin ha ye nzdik e ngoshtr psndid e shod /shod n gem PL EZ near EZ ring chose pp became.3SG /became 3PL the gems near the ring were chosen IPJ mobayl ha ye daxel e kif ha xamush shod /shod n cellphone PL EZ inside EZ bag PL o ff became.3SG /became 3PL the cellphones in the bags turned off. UPJ mobayl ha ye nzdik e kif ha xamush shod /shod n cellphone PL EZ near EZ bag PL Off became.3SG /became 3PL the cellphones near the bags turned off. USJ mobayl ha ye nzdik e kif xamush shod /shod n cellphone PL EZ near EZ bag Off became.3SG /became 3PL the cellphones near the bag turned off.

PAGE 169

169 Table 5 3 Test of within subject effect per participants Source D f Mean Square F S ig. S tructure 1 .125 2.290 .144 Error(structure) 23 .055 Prepositional phrase 2 .456 4.414 .018 Error(PP) 46 .103 structure prepositional phrase 2 .003 .076 .926 Error(structure*prepositioal phrase ) 46 .040

PAGE 170

170 Table 5 4 Te st of within subject effect ( per items ) Source df Mean Square F Sig. Prepositional phrase 2 .237 8.338 .001 Prepositional phrase structure 2 .002 .053 .948 Error(prepositional phrase) 44 .028

PAGE 171

171 Table 5 5 Mean proportion of mismatched ve rb usage with different modifying prepositional phrase Prepositional phrase Mean Std. Error Individual plural .255 .060 Unity Plural .406 .066 Unity singular .438 .059

PAGE 172

172 Table 5 6 Pairwise comparisons of proportion of mismatched verb usage w ith different modifying prepositional phrase (I) prepositional phrase (J) prepositional phrase Mean Difference (I J) Std. Error Sig. a Individual plural Unity plural .151 .071 .043 Unity singular .182 .076 .026 Unity plural Individual plural .151 .071 .043 Unity singular .031 .046 .503 Unity singular Individual plural .182 .076 .026 Unity plural .031 .046 .503

PAGE 173

173 Table 5 7 Mean proportion of singular verb usage (interaction of prepositional phrase and structure St ructure Prepositiona l Phrase Mean Std. Error Inchoative Individual Plural .219 .059 Unit Plural .385 .060 Unit Singular .406 .065 Passive Individual Plural .292 .073 Unit Plural .427 .082 Unit Singular .479 .070

PAGE 174

174 CHAPTER 6 EXPERIMENT 4 EFFECT OF SUBJECT N OUN CONCRETENESS Overview The goal of this study was to test the effect of Concreteness vs. Abstractness of inanimate plural subject nouns in optional subject verb number agreement in Persian. Eberhard (1999) investigated the effect of conceptual number of distributive subject phrases on verb agreement. The results indicated that the concreteness or imageability of the conceptual representation may affect the availability of conceptual number to the agreement process. Basically, more plural verbs were produ ced with concrete/ imageable distributive nouns (e.g. the stamp on the envelopes vs. the crime in the cities ). Similar to many other studies on agreement processing, Eberhard (1999) used the elicitation of agreement errors in a completion task. Optional subject verb number agreement in Persian provides a good situation in which to test the effect of Subject Concreteness vs. Abstractness on subject verb number agreement with normal grammatical sentences (because there is no need to use elicitation of agree ment errors or the ambiguity of distributive nouns). The experiment in this chapter asked whether the concreteness vs. abstractness of inanimate plural subject nouns cause Persian native speakers to produce more singular or plural verbs in optional subj ect verb number agreement. Based on an informal pilot study, participants were predicted to produce more mismatched (singular) verbs with the abstract inanimate plural subject nouns rather than with the concrete ones.

PAGE 175

175 Method Participants The participant s were twenty five Iranian native speakers of Persian (16 males and 9 females) between 22 30 years old (mean 26 years). They were members of University of Florida community, and participated in the study voluntarily. The participants were late learners of English who had learned this language for the purpose of education and did not have significant knowledge of any other language. All participants had normal or corrected vision and did not have any cognitive or neurological impairment. The participants of this experiment also participated in Experiments 1 (Chapter 3) and 3 (Chapter 5). Experiment 1 was conducted more than one year before Experiments 3 and 4, which were conducted at the same time. Materials Twenty four sets of target and seventy two distrac ter preambles were constructed. Twenty four items out of seventy two distracter items were the target preambles of Experiment 3 (see Chapter 5). The target preambles consisted of a concrete or an abstract inanimate plural subject noun plus either a past pa rticiple or an adjective, creating two similar structures, passive and inchoative auxiliary (see Chapter 1). In all the sentences, t he thematic role of the subject nouns was theme, so that all the subjects were acted on inanimate nouns. This experiment w as conducted in the same session as Experiment 3. Although the result of Experiment 2 (Chapter 4) showed similar effects of passive and inchoative auxiliary verbs on subject verb agreement in Persian (e.g. participants produced similar percentages of sing ular verbs with these two structures), the preambles of each structure were still coded differently. The code (V) was used for passive preambles and

PAGE 176

176 (J) for inchoative auxiliary (because inchoative consists of and adjective + the auxiliary /shod n/, to bec Also, d epending on the Concreteness vs. Abstractness of the subject nouns, two conditions were created and coded as (C) and (A) respectively. So, there were two factors: (1) Type of the subject (Concrete vs. Abstract subject head noun) and (2) Struc ture (Passive vs. Inchoative auxiliary). This created 4 conditions for the target sentences. A set of sample target preambles and possible responses of each condition is shown in T able 6 1 The participants were instructed to read the preambles and compl ete each sentence with their own verb, using informal, spoken Persian. Because of the passive and inchoative auxiliary structures used, giving the nonverbal part of the verbs meant that the participants just needed to complete the sentence with a specific auxiliary verb (/ shodn/ (to become). In order to have the participants use only the verb /shodn/ (to become) to get the desired structure, the participants were requested not to use the other possible option (the verb th adjectives, giving a copular structure). In order to keep the effect of tense constant, the participants were told to imagine that all the events of the sentences happened the day before, which allowed them to naturally use past tense verbs. The filler items included different kinds of structures. They consisted of a singular or plural subject noun, optional singular/plural (in)animate direct object noun, the direct object marker /ra/ (when there was a direct object), optional indirect object, and the nonverbal part of the compound verb. Using a Latin square design, the target sentences of each of the structures (passive or inchoative auxiliary ) were divided into two compatible experimental lists. In

PAGE 177

177 each list, there were twelve preambles of each cond ition (with concrete (C) vs. abstract (A) subject nouns). Six of the preambles were Inchoative auxiliary and the other six preambles were passive. No item repeated so that for each list, each participant saw every item exactly once, and across all the lis ts each item appeared once in each condition. The forty eight filler preambles were the same in both of the lists. The order of the trials was arranged pseudo randomly so that target sentences from the same condition did not appear next to each other in th e list. Each list was read by twelve participants. A list of the target and filler items in gloss and in Persian, along with the related conditions, are available in Appendices D at the end of this dissertation. Since understanding the literal meaning (th e third line of the gloss) could be difficult for non native speakers of Persian, an English grammatical equivale nt of the complete sentence was added in the forth line. Procedure Participants were tested individually. After giving informed consent, partic ipants sat in front of an Apple Mac computer equipped with a microphone running Psyscope X. The experiment started with instructions followed by four practice trials. As mentioned above, in order to avoid an effect of tense, the participants were told to i magine that all the events of the sentences happened the day before. When participants indicated they were ready, the experiment began. In each trial, the preamble appeared in black text at the center of the computer screen on a white background. The part icipants were instructed to repeat the preamble, complete the sentence, and then press any key to continue. As this point, a signal (++++++) appeared on the screen. The participants were supposed to press any key

PAGE 178

178 again when they were ready to proceed to th e next trial. After pressing the key, it would take one second until the next trial would appear. This would give enough time to mark separate file in the computer. This checklists. Design and Data Anal ysis The verbs produced by the participants were entered into an Excel spreadsheet, coded one or zero based on the singularity (1) or plurality (0) forms of the verbs respectively, giving us a proportion of singular (mismatching) verbs. Recall that the que stion for this experiment is whether there is an effect of Concreteness vs. Abstractness (of the subject noun) in the production of subject verb agreement. The dependent variable is the proportion of singular verbs out of all items. As already noted above, there were two independent, within subject factors: Structure (with two levels, Passive and Inchoative auxiliary ), and Type of the Subject Noun (with two levels, Concrete vs. Abstract subject noun). Sets of by participants (F1) and by items (F2) Repeate d Measures ANOVAs with the two level factor Type of the Subject and two level factor Structure were performed. In addition, planned paired sample t tests were performed. An alpha level of .05 was used for all statistical tests. excluded from analysis because the participant had difficulty performing the task in informal Persian language (see

PAGE 179

179 Results There were two independent, within subject factors: Struct ure (with two levels, passive and inchoative auxiliary) and Type of the Subject (with two levels, Concrete vs. Abstract). AJ: Abstract subject with inchoative verb, the realities became visible. CJ: Concrete subject with inchoative verb, the ships became visible. AV: Abstract subject with passive verb, the hopes were given. CV: Concrete subject with passive verb, the computers were given. This analysis revealed a main effect of Types of Subject (Concrete vs. Abstract) ( F 1 (1, 23) = 34.563, MSE = 0.046, p < 0.01; F 2 (1, 22) = 22.645, MSE = 0.035, p < 0.01). There was no effect of Structure (Fs < 1) or an interaction between Types of the Subject (Concrete vs. Abstract) and Structure ( F1 =1.960, F2 < 1). These results show that the Concreteness vs. Abstractne ss of the subject has a significant effect on the singularity or plurality of the verbs that the participants produced while Structure of the sentence does not The results are reflected in T ables 6 2 and 6 3. These results indicate that the Type of the s ubject (Concrete vs. Abstract) was a critical factor in optional subject verb agreement in Persian. This means that the participants produced more singular verbs with Abstract inanimate plural subjects compared to Concrete ones. Table 6 4 shows the mean pr oportion of mismatched verb used with Abstract and Con crete subject nouns and T able 6 5 shows the pairwise comparison of them. On the other hand, the lack of a significant effect of Type of the Structure suggests that the Passive and Inchoative auxiliary structures did not have different effects in optional subject verb agreement. Tables 6 6 and 6 7.

PAGE 180

180 Hence, the participants produced more mismatched (singular) verbs with the Abstract inanimate plural subject nouns either in passive or inchoative auxiliary structure (AV & AJ) compared to the Concrete inanimate plural subject nouns (CV & CJ). This result is shown in Table 6 8. Discussion The results of this study showed that the participants produced more mismatched verbs (singular verbs) with Abstract inanim ate plural subject nouns rather than with Concrete ones. This indicates that the Concreteness vs. Abstractness of the inanimate plural subject entities has a significant effect on the singularity/plurality of the optional agreement. But, the results of thi s study did not show any effect of the structure. This means that passive and inchoative auxiliary structures had almost the same effect on form of the verb (to produce singular or plural verb). This result is i n line with the results of Chapter 4, in whic h passive and inchoative auxiliary structure had almost the same effect in the production of optional subject verb number agreement). Crucially, the effect of Concreteness vs. Abstractness of the subject noun appears to be related to the semantic meaning r ather than grammatical meaning. As in Chapter 5, this implies the effect of semantic factors, rather than purely syntactic ones. Both abstract and concrete plural subject nouns have the same plural morpheme / ha/ but the participants produced more singular verbs with the Abstract plural subject nouns. This fact that the Abstractness of nouns could influence agreement is most consistent with a Maximalist view and constraint account in which additional, non syntactic factors may influence agreement.

PAGE 181

181 Conclusi on In this study, the Concreteness vs. Abstractness of inanimate plural subject nouns in the subject verb agreement was tested. Optional subject verb number agreement of Persian (for the inanimate plural subjects) provides an exceptional situation in which to test the effect of a semantic factor (Abstractness vs. Concreteness of the subject nouns) in subject verb agreement where all experimental materials are grammatical sentences with no need to use elicitation of errors or the ambiguity of collective or d istributive nouns. The preambles consisted of a subject head noun and non verbal part of the compound verb, which led me to use passive and inchoative auxiliary structures. I coded these two structures differently, leading to two factors: Structure (incho ative auxiliary vs. passive) and Type of the subject noun (Concrete vs. Abstract). As predicted, the results of the study showed that the singularity vs. plurality of the produced verbs was affected by the Concreteness vs. Abstractness of the subject noun s. More specifically, the participants produced more singular verbs with Abstract inanimate plural subject nouns rather than Concrete ones. This result, that optional subject verb number agreement is affected by a semantic factor (concreteness vs. abstrac tness of the subject noun) supports the maximalist constraint account which argues that in addition to syntactic information other factors including semantic factors influence agreement during grammatical encoding and that the agreement is not governed so lely by syntactic information.

PAGE 182

182 Table 6 1 S ample of preamble and possible response for concreteness experiment structure condition G loss possible target response inchoative abstract vaqe'iyt ha pdidar shod /shod n reality PL visible became.3SG /became 3PL the realities became visible. inchoative concrete keshti ha pdidar shod /shod n ship PL visible became.3SG /became 3P L the ships became visible. Passive abstract rftar ha psndid e shod /shod n behavior PL approved pp became.3SG /became 3PL the behaviors were approved. Passive concrete lebas ha psndid e shod /shod n clothe PL approved pp became.3SG /became 3PL the clothes were approved.

PAGE 183

183 T able 6 2 Test of within subject effect ( per participants ) S ources D f Mean Square F Sig. C oncreteness 1 1.584 34.563 .000 Error(concreteness) 23 .046 S tructure 1 .019 .687 .416 Error(structure) 23 .027 concreteness structure 1 .057 1.960 .175 E rror(concreteness*structure) 23 .029

PAGE 184

184 Table 6 3 Test of within subject effect ( per items ) Source df Mean Square F Sig. Concreteness 1 .792 22.645 .000 concreteness structure 1 .028 .811 .378 Error(concreteness) 22 .035

PAGE 185

185 Table 6 4 Mean proportion of mismatched verb with concrete and abstract subject Concreteness Mean Std. Error Abstract .705 .045 Concrete .448 .051

PAGE 186

186 Table 6 5 Pai rwise comparison of concrete vs. abstract subject (I) concreteness (J) concreteness Mean Diffe rence (I J) Std. Error Sig.a Abstract Concrete .257* .044 .000 Concrete Abstract .257* 044 .000

PAGE 187

187 Table 6 6 Mean proportion of singular verb produced in different structure Structure Mean Std. Error inchoative .590 .042 Passive .563 .050

PAGE 188

188 Table 6 7 Pairwise comparison of inchoative and passive structures (I) structure (J) structure Mean Difference (I J) Std. Error Sig.a inchoative P assive .028 .034 .416 P assive I nchoative .028 .034 .416

PAGE 189

189 Table 6 8 Mean proportion of singular verb usage (interaction of concreteness and structure) concreteness structure Mean Std. Error Abstract Inchoative .743 .050 Passive .667 .050 Concrete Inchoative .438 .048 Passive .458 .065

PAGE 190

190 CHAPTER 7 EXPERIMENT 5 OBJECT ATTRACTION EFFECT Overview As discussed in Chapter 1, in Persian, singular nouns are the unmarked form and thus do not have any overt morpheme to show singularity. On the other hand, plural nouns may get different plural marking morphemes depending on the animacy of the noun. Persian employs two plural morphemes: / ha/ and / an/. The suffix / ha/ can be used for all plural nouns regardless of animacy while the suffix / an/ is used only for animate plural nouns. One of the goals of this study is to test the effect of m orphology (that is, the effect of different plural morphemes) in optional subject verb agreement. (7.1) z n ha chtr ha woman PL umbrella PL z n an *chtr an woman PL umbrella PL As discuss ed in Chapter 2, Bock & Miller (1991) and the replications of it in several SVO languages have reported an attraction effect of local nouns ( i.e. the argument NP of a prepositional phrase on the head noun subject (e.g., the key to the cabinets ) on subject verb agreement. However, while the results of these studies have been very informative, they are still somewhat problematic in two ways: first, because

PAGE 191

191 an attractor remai ns unclear and, second, there remains an ongoing debate about whether object nouns can serve as attractors. Some experiments have tested the attraction effect of the intermediary object pronoun in SVO languages. Antn ed the attraction effect of the preverbal, pronominal direct object, however, the results showed no attraction effect. This finding could be due to the effect of syntactic function in that the object is out of subject phrase and due to insulation of syntax it cannot affect the subject verb agreement. There could also be an effect of the part of speech. That is, the effect of the object noun could be different from the object pronoun because the processing of pronouns could be different from words as the pr onouns may have accusative markers. This is the case in Spanish. The unmarked word order is SVO but the object pronoun, which has accusative case, is preverbal. On the other side, Hartsuiker, Antn Mndez, & van Zee (2001) tested the attraction effect of i ntermediary object nouns in complement clauses in Dutch. In Dutch, as the following example shows, the object of the complement clause is preverbal. (7.2) Karin zegt dat het meisje de krans en WIN Karen says that the girl the garland s WIN The results showed the attraction effect of the object (of the embedded clause), but the effect was less than the attraction effect of the subject modifier phrase. The same authors also tested the effect of obje ct pronouns in the embedded clause in two conditions: case ambiguous pronouns vs. unambiguous pronouns. The results showed

PAGE 192

192 a strong attraction effect of embedded clause object pronouns when case was ambiguous. But, the unambiguous objects did not show att raction effect. The Persian language has some specific characteristics that make it a special language for studying agreement. The unmarked word order in Persian is Subject Object Verb (SOV language), and thus the object noun naturally intervenes between the subject and its verb, creating a good test case for object attraction without needing to rely on marked syntactic structures or clitics (e.g. in French by Hartsuiker, Antn Mndez, & van Zee (2001)). (7.3 ) les pommes, je les ai manges the apple s, (fem, pl) I them (object, clitic) have eaten (fem, pl) Second, and more fundamentally, because all of the results thus far have obained in languages in which number agreement is always required, they are based on the pr oduction and comprehension of ungrammatical structures. While this approach has provided valuable data and insights into how grammatical information is used during online processing, it is less than ideal for data to come solely from studies that require p articipants to either produce or read ungrammatical structures. Optional subject verb number agreement in Persian (third person inanimate plural subject nouns can have singular or plural verbs) provides the opportunity to have grammatical acceptable senten ces with matched (plural) and mismatched verbs (singular). Thus, the main goal of this study was to test the influence of the main clause object noun on the optional subject verb agreement in an SOV language, and more specifically, to test the effects of animacy, number, and different types of plural

PAGE 193

193 morphemes of the object nouns in the singular vs. plural form of the produced verb Considering the intermediary position of the object in SOV order sentences in Persian, I predict that the object will serve a s an attractor for number marking on the verb: More plural verb morphology is expected when the object is plural compared to when it is not. Further, because subject verb agreement for animate subjects is obligatory in Persian but not for inanimate plural subjects, the animacy of the object may influence the strength of any attraction effect, with strongest effects occurring for animate objects (the obligatory case for subject agreement). The predictions for the different types of plural morpheme are less c lear, but given that the plural morpheme / ha/ is the unmarked suffix used for all animate and inanimate nouns while the plural morpheme / an/ is specifically used for animate nouns, we can expect that speakers will produce more plural forms with an animat e plural objects that are suffixed with / an/ compared to / ha/. Methods Participants The participants were 46 Iranian native speakers of Persian (29 males and 17 females) between 22 43 years old (mean 28 years). They were members of the University of Fl orida community, and participated in the study voluntarily. The participants were late learners of English who had learned this language for the purpose of education and did not have significant knowledge of any other language. All participants had normal or corrected vision and did not have any cognitive or neurological impairment. The participants of this experiment also participated in Experiment 2 (Effect of Verb Type & Verb Tense), which was conducted at the same time.

PAGE 194

194 Materials Forty target and forty distracter items were constructed. The target items consisted of a sentence preamble with an inanimate plural subject noun, a direct object noun, the direct object marker /ra/ (seen as /ro/ or /o/ in spoken language) and the nonverbal part of a transitive compound verb. The thematic role of the subject was used to accomplish a job, and were also the subject of the sentence in the absence of the agent noun. In order to av oid effects of subject concreteness/abstractness, only concrete nouns were selected as subject of the sentence. By systematically varying the animacy, number, and plural suffix type of the object nouns, five conditions of preambles were created: 1) APN: an imate plural object with suffix / an / (e.g. sb an, horse ) 2) APH: animate plural object with suffix / ha / (e.g. sb ha, horse ) 3) ASO: animate singular object with no suffix (e.g. ) 4) IPH: inanimate plural object with suffix / ha / (e.g. mojseme ha, statue PL, ) 5) ISO: inanimate singular object with no suffix (e.g. mojseme, statue.SG, ) All the verbs used in this experiment were combined compound verbs, and as such, participants were ins tructed to read the preambles out loud and then create a complete sentence using informal, spoken Persian. By providing the nonverbal part of the compound verbs in the preamble, participants crucially only needed and were restricted to completing the sente nce with the verbal part of the compound verb as can be seen in Table 7 1. In order to grammatically and sensibly complete the preambles,

PAGE 195

195 participants needed to use the verbal part /krdn/ (to do) in twenty nine of the items, /dadn/ (to give) in five, / bordn/ (to take) in three, /dashtn/ (to have) in two, and /gereftn/ (to take) in one. A set of sample target preambles for one item and possible responses of each condition is shown in Table 7 1. As we saw in Chapter 4, verb tense appears to interact wi th subject verb agreement phenomena in Persian. Thus, in order to control for this effect and keep it constant across trials, participants were told to imagine that all the events in the sentences happened the day before, which allowed them to naturally us e the past tense verbs. The filler preambles looked superficially like the target preambles, but included different kinds of structures. These filler preambles consisted of a singular or plural subject noun, singular/plural (in)animate object noun, the di rect object marker /ra/, and the nonverbal component of a transitive complex predicate. The purpose of these filler items was to prevent expectations and boredom in the participants. Using a Latin square design, the target items were divided into five comp atible experimental lists. In each list, there were eight items of each condition (with APN, APH, ASO, IPH, and ISO objects), with no item repeated within a list, so that for each list, each participant saw every item exactly once, and across all the lists each item appeared once in each condition. The forty filler items were the same in all the five lists. The order of the trials was arranged pseudo randomly so that target sentences from the same condition did not appear next to the each other in the list. In total, each list was read by nine participants.

PAGE 196

196 A list of the target and filler items in gloss and in Persian, along with the related conditions, are available in Appendix E. Since understanding the literal meaning (the third line of the gloss) could be difficult for non native speakers of Persian, an English grammatical equivale nt of the complete sentence was added in the forth line. Specific and unspecific direct object in Persian In this experiment, I tested the effect of intervening object noun in the optional subject verb agreement. The object nouns of the target preambles were [+definte, +specific] and followed by direct object marker (OM) /ra/. Considering the probable effect of object marker /ra/ in distinguishing the object argument from the r est of the sentence, the question may be raised as why to use definite objects which have the marker /ra/ and not objects without object markers. To answer this question, I need to explain the types of direct objects with respect to specificity and definit eness. Manipulating specificity and definiteness in Persian gives us three kinds of nouns: [ specific, definite], [+ specific, definite], and [+ specific, + definite]. Objects that are [ specific, definite] are always singular they cannot be plura l. Neither can they have a demonstrative or possessive pronoun nor be followed by the object marker /ra/. As was explained in Chapter 1 (c.f. Incorporation and example 1.18), the direct objects of these kind of sentences lose their grammatical endings and incorporate with the verb. In this way, the non specific object and the transitive verb these incorporated compound verbs is transparent (Dabir Moghaddam, 1997).

PAGE 197

197 (7.4) (Niki) sib chid (Niki) apple picked.3SG Lit., (Niki) did apple picking Therefore, it was not possible to use [ specific, definite] objects in this experiment, because this type of object incorporates with the verb, ultimately becom ing a part of the compound verb. Moreover, it does not have plural form; so, testing the attraction effect of the object and its features in the agreement was not possible. Objects that are [+specific, definite] can be singular or plural, but cannot hav e demonstrative or possessive pronouns. They do, however, need the indefinite marker / i/ with singular and plural nouns or the number of the items before the name. This type of object does not need to receive the object marker /ra/. (7.5) a. (Niki) sib i chid (Niki) apple INDEF picked.3SG b. (Niki) yek sib (i) chid (Niki) one apple (INDEF) picked.3SG c. (Niki) sib ha i chid (Niki) one apple INDEF picked.3SG

PAGE 198

198 d. (Niki) chnd ta sib chid (Niki) a few apple picked.3SG Thus, I did not use [+specific, definite] object nouns because the frequency of this type of object (especially with animate plural noun with suffix / an/) did not seem to be common e nough. The examples (7.5a) and (7.5c) are grammatical and acceptable sentences but it is difficult to create enough simple target sentences of this kind (i.e. with inanimate plural subject nouns). Moreover, the indefinite morpheme / i/ is suffixed to the s ingular noun or to the plural suffix / ha/ or / an/. Hence, the indefinite marker / i/ intervenes between the object noun and the verb. Examples (7.5b) and (7.5d) with number or quantifiers and a singular form of the noun seem to be more common. But, the s ingularity or plurality of nouns is shown by the number or quantifiers preceded the nouns and not plural suffix / ha/ or / an/. Conversely, [+definite, +specific] d irect object nouns in Persian are followed by the marker /ra/. These nouns can be singular or plural and have demonstrative and possessive pronouns. (7.6) a. (Niki) sib ro chid (Niki) apple OM picked.3SG b. (Niki) sib ha ro chid (Niki) apple PL OM picked.3SG

PAGE 199

199 c. (Niki) sib esh ro chid (N iki) apple her OM picked.3SG d. (Niki) un sib ro chid (n in yeki) (Niki) that apple OM picked.3SG (not this one) I used this type of object because I could manipulate the animacy, num ber of the object noun, and type of the plural morphemes to create more natural target preambles. Procedure Participants were tested individually. After giving informed consent, participants sat in front of an Apple MacBook Pro computer equipped with a m icrophone and running Psyscope X. The experiment started with a training session including instructions and seven trials. As mentioned above, in order to avoid an effect of tense, the participants were told to imagine that all the events of the sentences h appened the day before. When participants indicated that they were ready, the experiment began. In each trial, the preamble appeared in black text at the center of the computer screen on a white background. The participants were instructed to repeat the p reamble and then complete the sentence aloud. Next, they pressed a key to continue to the following screen, which displayed a signal (++++++) indicating that participants could pause on this screen until pressing a key to continue to the next trial. After pressing the key, there was a one

PAGE 200

200 recorded in a separate file in the computer. This recording was consulte d whenever the Design and Data Analysis The verbs produced by the participants were entered into an Excel spreadsheet and coded as one or zero based on the singularity ( 1) or plurality (0) of the verbs. This gives us a proportion of the trials with grammatical but mismatching singular verb use. Recall that the key question for this experiment is whether there is an effect of the intervening object argument in the producti on of subject verb agreement. The dependent variable is the proportion of singular verbs out of all items. As already noted above, there were three independent, within subject factors: Animacy (with two levels, animate vs. inanimate), Number (with two leve ls, plural with the suffix / ha/ vs. singular), and Type of Plural Morpheme used for the animate subjects (with two levels, animate plural wi th the suffix / an/ vs. animate plural with the suffix / ha/). Sets of Repeated Measures ANOVAs for each morpheme type (/ ha/ and / an/) were performed, with Object Animacy (animate vs. inanimate) and Object Number (singular vs. plural) as independent variables and participants (F1) and items (F2) as random variables. In addition, planned paired sample t tests were c onducted. An alpha analysis because the participant had difficulty performing the task in a timely fashion (she had been living in another state for two years before moving to Florida and had not were used in the analysis.

PAGE 201

201 Results First, an analysis of / ha/ was conducted, with the following four conditions: APH, ASO, IPH and ISO, su ch that there were two independent factors, Object Animacy (Animate vs. Inanimate) and Object Number (Plural vs. Singular). APH: animate plural object with suffix / ha / (e.g. sb ha, horse ) ASO: animate singular object with no suffix (e.g. s ) IPH: inanimate plural object with suffix / ha / (e.g. mojseme ha, statue PL, ) ISO: inanimate singular object with no suffix (e.g. mojseme, statue.SG, ) This analysis revealed a main effect of Object Number ( F 1 (1, 44) = 5,401, MSE = 0.020, p < 0.05; F 2 (1, 39) = 0.4,291, MSE = 0.022, p < 0.05), with participants producing significantly more singular verbs when the object was a singular noun. There was no effect of Animacy (F1 < 1, F2 < 1) nor an interaction between An imacy and Number ( F1 = 1.928, F2 < 1). This shows that the animacy of the intervening object nouns did not have any effect on the production of singular or plural form of the verbs that the participants produced. These results are reflected in Table s 7 2 a nd 7 3. These results indicate that the Number feature of the object noun produces an attraction effect in optional subject verb agreement in Persian and that under nearly equal conditions with inanimate plural subjects, speakers produced more singular ver bs when the objects of the sentences were singular rather than plural and marked with the suffix / ha/. In other words, the participants used fewer singular verbs in APH and IPH conditions, compared to ASO and ISO conditions. On the other hand, the lack o f a

PAGE 202

202 significant effect of Animacy suggests that the animacy of the object noun alone is not a factor in optional subject verb agreement. Next, the following four conditions (ASO, ISO, APN, IPH) were compared, with that same two independent variables as be fore, but with / an/ instead of / ha/. APN: animate plural object with suffix / an / (e.g. sb an, horse ) ASO: animate singular object with no suffix (e.g. ) IPH: inanimate plural object with suffix / ha / (e.g. mojseme h a, statue PL, ) ISO: inanimate singular object with no suffix (e.g. mojseme, statue.SG, ) In this case, the analysis again revealed a significant main effect of Object Number ( F 1 (1, 44) = 15,754, MSE=0.021, p< 0.01; F2 (1, 39) = 14,421, MSE= 0.021, p<0.01) showing that participants produced more singular verbs with inanimate plural subjects when the objects were singular. Additionally, while there was again no main effect of Animacy ( F1 < 1, F2 = 1.519 ), indicating that animate and inan imate object nouns had the same impact on participants' use of singular and plural verbs overall, there was, however, a significant interaction of Animacy and Number ( F1 (1, 44) = 7,366, MSE=0.023, p< 0.01; F2 (1, 39) = 4,586, MSE = 0.033, p<0.05), unlike in the case of / verbs when the object was an animate plural noun with suffix / an/ (e.g. sb an, horse ) compared to when it was an inanimate singular noun (e .g. mojseme, ). There was not such a meaningful interaction of Animacy and Number in the previous comparison, when the APH (animate plural with suffix / ha/) was

PAGE 203

203 among the conditions. This result shows the effect of morphology the diff erence between APN (animate plural with suffix / an/) and APH (animate plural with suffix / ha/). These results are reflected in Tables 7 4 and 7 5. The interaction of Animacy and Number is more evident in Table 7 6, in which the proportion of singular ver b usage broken down by condition is shown. The proportion of singular verb usage was much higher for APH (0.24) rather than with APN (0.16). This is the main effect of morphology (using different plural suffix for animate plural subjects in almost equal co ndition). A pairwise comparison of plural morphemes / ha/ and / an/ was conducted and the results showed that the difference was significant. APH t(1, 44) = 7.248, P < 0.01. APH t(1, 44) = 5,965, P < 0.01. The results are shown in Table 7 7. Discussion The se results indicate that the number of object nouns plays a significant role in the selection of singular vs. plural forms of the verbs. But, while the animacy of the subject influences subject verb agreement patterns in Persian, this experiment shows that the animacy of the object does not have a significant impact on the selection of the singular or plural forms of the verb. Further, comparing the results of plural morpheme markers / ha/ and / an/, and in particular the interaction of animacy and number a bove present for / an/ marked plurals but not for / ha/ marked plurals, indicates that the attraction effect of plural animate objects with suffix / ha/ (APH) are different from the attraction effect of plural animate objects with suffix / an/ (APN). This suggests that animate plural objects with suffix / ha/ do not cause attraction for subject verb agreement. However, animate plural objects with the suffix / an/ were shown to cause a significant attraction effect for agreement: in nearly identical conditi ons, the participants

PAGE 204

204 produced less singular verbs with inanimate plural subjects when the object was animate plural with suffix / an/ than when the object was animate plural with suffix / ha/. This finding suggests an important effect of morphology (usin g a different plural suffix for intervening animate plural object nouns) in the optional subject verb agreement in Persian. Crucially, the effects seen here appear to be morphological in nature rather than purely syntactic. The animate plural object nouns with the suffixes / ha/ and / an/ are syntactically the same but morphologically different, with the plural suffix / ha/ being unmarked and used for all types of plural nouns while the plural suffix / an/ used only for animate nouns. In fact there are two points. First, this study shows the attraction effect of the number feature of an object noun as a non subject argument in subject verb agreement. Second, the study shows the attraction effect of morphology (specific type of plural morpheme used for anima te nouns). These effects are not purely syntactic. That is, both syntactic information (plurality of object noun) and morphological information (animacy) of the object noun and the effect of the specific plural morpheme are involved. The fact that these pa tterns in morphological marking could influence agreement is most consistent with the Constraint based account in which additional, non syntactic factors like animacy and patterns of co occurrence with animacy (such as the case of / an/) of a non subject a rgument may influence agreement. As explained in Chapter 2, one of the advantages of the Constraint account is its ability to explain the attraction effect in the same framework as it does for normal which the semantic and grammatical constraints happen within the normal range for

PAGE 205

205 agreement (based on the meaning, structural distances, and morphological cues). But, variations such as attraction happens if the semantic, st ructural, or morphological constraints are outside the normal, usual range of values. The results of this experiment cannot be explained by Minimalist approach and the more recent Control account as these accounts give a special role to syntactic propertie number feature from the message. Then, the number feature of the subject is transmitted to the ve rb and there is no other syntactic mediation between them. This means that the subject number controls the verb number. So, the subject is the agreement controller and the verb is the agreement target. Because these approaches encapsulate the subject from other syntactic features of the sentence but outside the subject phrase, they cannot explain the attraction effect of phrases outside of the subject phrase on the subject verb agreement. So, they cannot explain the attraction effect of direct object nouns in the optional subject verb number agreement evidenced in Persian. According to the maximalist view, each of the two elements (e.g. subject and verb) involving in the agreement relation express some information about the relation. Through unification, th e information of these two involving elements is merged and create the agreement relation. Importantly, unification of features can explain the result of this experiment. Based on this view, features are unified (they are not copied or transported from one element to another) and, as a result, they can be shared by elements of different branches of the tree structure. This point is particularly useful

PAGE 206

206 when thinking about subject verb agreement in Persian because it involves syntactic and non syntactic featu res of involving and non involving elements of the agreement. Conclusion In this study, the attraction effect of the object noun in subject verb agreement was tested. The SOV word order of Persian and its optional subject verb number agreement (for the ina nimate plural subject nouns) created an exceptional condition to test the effect of intermediary object arguments in the optional subject verb agreement (where all the materials were grammatical sentences and there was no need to use elicitation of errors) There were three factors: Number, Animacy, and Morphology. As was predicted, the results of the study showed that the singularity vs. plurality of the produced verbs was affected by the singularity vs. plurality of the intermediary objects. More specific ally, the participants produced more singular verbs when the objects were singula r and more plural verbs when the intervening objects were plural. But, contrary to the predictions, the results did not show an attraction effect for object animacy: participa nts produced a similar range of singular verbs when the objects were singular animate /inanimate nouns as when they were plural animate/inanimate nouns with the unspecific suffix / ha/. Finally, the experiment showed an effect of morphology: participants pr oduced fewer singular verbs when the local animate plural object nouns were suffixed with the animate specific plural morpheme / an/ rather than with the unspecific plural morpheme/ ha/. The first and final result the attraction effect of object noun; the effect of number and morphology (because of the specific type of morpheme) support the Constraint account which argues that semantic, morphological, and syntactic information influence

PAGE 207

207 agreement during grammatical encoding and the agreement is not governe d solely by syntactic information. Also, the subject verb agreement can be affected by the attraction effect of a noun phrase/local noun outside the subject phrase.

PAGE 208

208 Table 7 1 S ample preamble and possible response for each condition Condition Preamble P ossible response APN wagon ha sb an R o Hml ...... wagon ha sb an ro hml Krd/ krd n. wagon Pl horse Pl OM carrying wagon Pl horse Pl OM carrying did.3SG / did 3PL APH wagon ha sb ha R o Hml ...... wagon ha sb ha ro hml Krd/ krd n. wagon Pl horse Pl OM carrying wagon Pl horse Pl OM carrying did.3SG / did 3PL ASO wagon ha sb R o Hml ...... wagon ha sb ro hml Krd/ krd n. wagon Pl horse OM carrying wagon Pl horse OM carrying did.3SG / did 3PL IPH wagon ha mojseme ha R o Hml ...... wagon ha mojseme ha ro hml Krd/ krd n. wagon Pl statue Pl OM carrying wagon Pl statue Pl OM carrying did.3SG / did 3PL ISO wagon ha mojseme R o Hml ...... wagon ha mojseme ro hml Krd/ krd n. wagon Pl Statue OM carrying wagon Pl statue OM carrying did.3SG / did 3PL

PAGE 209

209 Table 7 2 Test of within subject effect per participants (conditions: APH, ASO, IPH, ISO) Source Df Mean Square F Sig. A nimacy 1 .007 .218 .643 Error(animacy) 44 .032 Number 1 .106 5.401 .025 Error(number) 44 .020 anima cy number 1 .025 1.928 .172 Error(animacy*number) 44 .013

PAGE 210

210 Table 7 3 Test of within subject effect per items (conditions: APH, ASO, IPH, ISO) Source D f Mean Square F Sig. A nimacy 1 .006 .315 .578 Error(animacy) 39 .020 N umber 1 .095 4.291 .04 5 Error(number) 39 .022 animacy number 1 .022 .844 .364 Error(animacy*number) 39 .026

PAGE 211

211 Table 7 4 Test of within subject effect per participants (conditions: APN, ASO, IPH, ISO) Source df Mean Square F Sig. Animacy 1 .028 .975 .329 Error(a nimacy) 44 .029 Number 1 .334 15.754 .000 Error(number) 44 .021 animacy number 1 .168 7.366 .009 Error(animacy*number) 44 .023

PAGE 212

212 Table 7 5 Test of within subject effect per items (conditions: APN, ASO, IPH, ISO) Source Df Mean Square F Sig. A nimacy 1 .025 1.519 .225 Error(animacy) 39 .016 N umber 1 .297 14.421 .000 Error(number) 39 .021 animacy number 1 .149 4.586 .039 Error(animacy*number) 39 .033

PAGE 213

213 Table 7 6 Mean proportion of singular verb usage with differen t verb groups Condition Number Number Marking Animacy Mean StErr APN Plural / an/ Animate .164 .034 APH Plural / ha/ Animate .239 .033 ASO Singular Animate .311 .034 IPH Plural / ha/ Inanimate .250 .035 ISO Singular Inanimate .275 .031

PAGE 214

214 Table 7 7 Pairwise comparison of plural morphemes / ha/ and / an/ t df Sig. (2 tailed) Mean Std. Deviation Std. Error Mean Pair 1 APH APN .07500 .21716 .03237 2.317 44 .025

PAGE 215

215 CHAPTER 8 CONCLUSION AND DISCUSSION Overview This dissertation presents a psycholinguistic study of subject verb number agreement production in Persian. As previously discussed, verbs in Persian, similar to many other languages, conjugate for three persons and two numbers, singular and plural. So, there is subject verb number an d person agreement in this language. But, Persian has an exception: subject verb number agreement is optional when the subject of the sentence is an inanimate plural noun. In this case, the related verb can be either singular or plural. While Persian gramm arians and linguists have suggested that factors such as the conceptualization of subject nouns, verb tense or type of the verb could be influential in the selection of the singular or plural form of the verb, no systematic study has previously tested the effects of these potential factors in agreement. Additionally, while the factors that influence subject verb number agreement have been tested experimentally in other languages, Persian has a number of characteristics that make it an informative case: Fir st, subject verb agreement studies in other languages have necessarily required the elicitation of errors and attraction effect techniques to test the effect of syntactic and non syntactic factors in agreement. Thus, the stimuli in these experiments were s ometimes ambiguous sentences and the analyzed data were ungrammatical sentences with agreement errors. The optionality of subject verb number agreement in Persian offers an excellent opportunity to study subject verb number agreement using grammatical Pers ian sentences. There is no need to elicit agreement errors and use ungrammatical sentences and agreement errors as

PAGE 216

216 data because both matched (plural form) and mismatched (singular form) verbs are grammatical and acceptable. The second useful characteristic for Persian concerns the type of task that can be used: previous studies in agreement have used a completion task in which participants were provided with a series of sentence preambles that they repeated before completing with their own words. Participan ts were free to complete the sentence with any verb/structure, but using different verbs/structures could be an interfering factor in the experiments, affecting agreement errors. Using compound verbs of Persian gives us a golden opportunity to avoid this p otential nuisance variable. In my experiments, the non verbal part of compound verbs (as a part of the preambles) were given to the participants so that they were unknowingly restricted to using a specific verb to complete the preamble. Thus, I was able to specifically test the effect of verb type in agreement and keep the effect of verb type constant across experiments. Third, previous studies have tested the effect of morphology in agreement using irregular/invariant plural nouns, but sometimes because of the properties of the language they could not test this factor alone (i.e. they had to mix this factor with another factor such as gender). Again, Persian has a property which can be very helpful to test the effect of morphology. Having two plural morphem es, / ha/ (for all nouns) and / an/ (specifically for animate nouns), provided an excellent opportunity to test the effect of morphology in agreement. Finally, previous studies have tested attraction effects of local nouns, but in most of these experiment s, the local noun was a noun phrase of a prepositional phrase modifying the subject head noun (thus, the local noun was a part of the subject phrase).

PAGE 217

217 In some experiments, the local noun was a noun phrase in a subject modifying relative clause. In a few ot her experiments in languages like French, the local noun was an object clitic pronoun (not noun) located between the subject noun and the verb. In fact, the word order of the studied languages in these experiments was SVO and thus researchers could not tes t the intervening attraction effect of an object noun as a local noun. Persian, like any SOV language, can be useful because the word order of unmarked sentences is SOV, and so the object falls naturally between the subject and verb. This provides a way to test the attraction effect of objects in optional subject verb agreement. The goal of this dissertation was to test the effect of probable factors in optional subject verb number agreement in Persian. This was accomplished through five experiments presen ted in Chapters 3 7. In this concluding chapter, I will now summarize and discuss the results of these experiments, and then discuss the bigger picture that these results provide with regard to our understanding of subject verb agreement. Experiment 1 The first experiment, (presented in Chapter 3) had two main goals and an ancillary goal: First, the acceptability of sentences with matched (plural) and mismatched (singular) verbs was tested. More specifically, the effect of Verb Type, Verb Tense, and Verb Nu mber in the acceptability of sentences with mismatched verbs was tested. For the second goal, the relationship between the acceptability of a mismatched verb form with the perceived thematic role of the subject (role interpretation) was tested. Again, effe cts of Verb Type, Verb Tense, and Verb Number were tested. Finally, the materials of this experiment were the complete versions of the sentence preambles used in Experiment 2 (Chapter 4), which allows us to compare acceptability with production.

PAGE 218

218 First, ve rb tense did not show any significant effect on the acceptability of number mismatch or the interpretation of the role of the subject nouns. Acceptability showed the effect of verb number in accepting sentences with matched and mismatched verbs. All the s entences with matched verbs in different groups were accepted but a different proportion of mismatched verbs were accepted for different verb groups. This is consistent with the effect of verb number in role interpretation; patient role s are more likely to be perceived for the subject s of sentences with mismatched verbs across different verb groups. For the effect of verb type, the results of role interpretation did show a significant effect of verb type: verbs of different types caused participants to inte rpret different thematic roles for the subject nouns of the sentences. These results allowed me to subjects, (2) verb types B, C & D with agent like subjects, (3) verb type E with agent subjects, and (4) verb types F & H with patient subject. Analysis of the acceptability rates, matched against these role interpretation results, showed that all of the sentences with plural (i.e. matched verbs) were accepted. Analys is of the mismatched verbs showed that the type of verb had a significant effect on ore than 90% of the sentences with mismatched (singular) verbs were judged acceptable with verbs in which subjects had a patient thematic ro almost no accepted mismatched verbs with verbs that assign agent thematic roles to their subjects. An interesting in between category, verbs of emission and verbs of instrum ent, had high rates of acceptability (75% 80%).

PAGE 219

219 These results suggest that when plural subjects have the role of agent, comprehenders disprefer agreement mismatch, and conversely, accept more mismatched verbs with patient or patient like subjects. They ar e also more likely to perceive patient or patient like thematic roles for the subject of sentences with a mismatched verb. These results thus show that the acceptance of sentences with mismatched verbs is much easier if the verbs of the sentences assign p atient, patient like or no role to their subject nouns rather assigning agent or agent like roles, establishing that subject thematic role affects subject verb agreement in Persian. There may actually be historical reasons for the effect of the effect of verb type and its thematic role. In classic Persian, only singular verbs were used for the inanimate plural subject noun unless the nouns were used metaphorically and were personified for an action that only animate beings could do. With the development of modern technology, inanimate objects now appear to do many activities on their own. So, in our his is especially true for transportation instruments. (8.1) The elevator went up. (8.2) The helicopter turned around. (8.3) The computer works fast. So, the speaker and comprehenders may consider them as the agent of the sentences, either because the re is no agent for the action or the role of the agent is not reflected in the sentence.

PAGE 220

220 The same is true for the subjects of verbs of emission. Although the subject nouns do not voluntarily do the action, there is no other agent implied in the activity. So, the reader and comprehenders may consider them to be the agent of the sentence and thus do not accept the mismatched (singular) form of the verb for them. (8.4) The lamps gave light. (8.5) The glasses gleamed. (8.6) The electrical cords spark. Det ermining an agent role for the subjects of verbs of emission, verbs of instrument, and even verbs of entity specific inchoative verbs (e.g. below ) is not a conceptualization of may think that they are not doing the action by themselves (e.g. for subjects of instrument verbs) or think that they are not doing any action (it is in their nature). (8.7) The iron rusted. (8.8) The seeds grow. The results of Experiment 1, showing that participants accept a mismatched (singular) form of the verb in optional subject verb number agreement depending on the thematic role that the verb assigns to its subject, demonstrating the impact of a semantic factor and thus are consistent with the maximalist approach and constraint based view in which semantic factors may influence agreement during grammatical encoding. E xperiment 2 As was discussed in Chapter 4, the effects of verb type and verb tense in the production of optional subject verb number agreement were tested in a completion task experiment. So, there were two factors: Verb Type (in eight groups) and Verb Tense

PAGE 221

221 (present vs. past) Participants were requested to read a series of preambles in two lists and complete the sentences with their own verbs imagining that the events happened the day before or on the same day. The materials of this experiment were preambles consisting of the subject head noun and non verbal part of eight different types of combined compound verb (i.e. the beginning part of the complete sentences used in Experiment 1). Similar to Experiment 1, the results showed a significant effect of verb type: the participants produced the highest proportion of mismatc hed verbs with verb groups that assign patient roles or no role to their subject nouns. More specifically, the participants produced almost similar portion of mismatched (singular) verb (32% to 38%) with the verbs in groups (A, B, F, and H) with patient su bjects and group G (with stative auxiliary verbs from the other verb groups that had subjects with roles other than patient: 17% These results show that the type of verb, and in particular the thematic role assigned to subjects influences the production of o ptional subject verb agreement. This result is fully consistent with the results from Experiment 1: the participants in Experiment 2 produced more mismatched verbs in sentences with patient subject compared to sentences with subjects of verbs of emission o r instrument, and produced no mismatches for verbs of agency. This shows that the participants prefer to produce mismatched verbs only for non agent subjects. The results of Experiment 1 showed the same effect of verb type: participants accepted more misma tched verbs in sentences

PAGE 222

222 with patient subjects and none or almost none in sentences with agent or agent like subjects. Also, they perceived more patient thematic role for the subject of sentences with mismatched verbs. Taken together, these results show a remarkable consistency between the influence of thematic role in the production and comprehension of subject verb agreement. One difference between Experiments 1 and 2 is that verb tense had a significant effect in the production of optional agreement in E xperiment 2: participants produced more mismatched verbs in the past tense compared to the present tense. However, in Experiment 1 the (present or past) tense of the verbs did not cause participants to accept more sentences with mismatched verbs in any ver b group, nor did it cause participants to perceive more patient role for the subjects of any verb groups. But, these results are not necessarily conflicting. For example, it could be possible that the effect of the verb tense is not as strong as verb type, or that its influence is limited to production. However, it is important to realize that while the effect of verb tense is only seen in production of present, this does n ot signal that participants would actually reject present tense sentences with mismatched verbs or would perceive different roles for subject nouns in sentences with present or past tense with mismatched verbs. That is, while Persian speakers are affected by verb tense to produce more mismatched verbs with past tense, the data from Experiment 2 does not show that speakers reject present tense sentences with mismatched verbs, simply that they do not tend to produce them. While it is not clear why verb tense has the effect it does in the production of autonomy in

PAGE 223

223 This means that the entities of the inanimate pl ural subject noun might seem to be more agent like and have control on their action when the tense of the verb is not past. On the contrary, they are considered the acted on of the events when the verb is past tense and the event is finished. The results o f role interpretation did not show this effect; the participants did not indicate more patient roles for subjects of past tense sentences compared to present tense ones. It could be possible that entities of subject nouns are conceptualized as individuals in sentences with present or future tense but as a unit in sentences with present and past tense verb is recommended. So, the results of Experiment 2 are consistent with those from Experiment 1: Participants produced more mismatched forms of the verb in the sentences that they were mostly to rate as acceptable with mismatched. Together, they strongly suggest that participants prefer mismatched verbs only for non agent sub jects. Again, these results are compatible with the view that in classic Persian, only singular verbs were used with inanimate plural subject nouns unless the nouns were used metaphorically and they were personified for an action that only animate beings could do. But, today, it seems that the inanimate objects do some activities by their own. This is reflected in our language. They are not instrument (e.g. they turned around by the helicopters ) but they are intermediary instruments used as subject of the sentence (e.g. the helicopters turned around). This implies that the inanimate subjects are considered as agent of the action. Persian speakers have generalized this for the subjects of verbs of emission ( e.g., The lamps give light. or The glasses are glea ming ). Although the

PAGE 224

224 subject nouns do not voluntarily perform the action, Persian speakers consider them an agent of the sentence and prefer not to use mismatched (singular) verbs for them. Again, similar to the judgment task, I believe that Persian speake rs use the mismatched form of the verb for the inanimate subject nouns whose thematic role is not f the doing the action by themselves (for subjects of instrument verbs) or think that they are not doing the action or the action is in their nature (the verbs of emiss ion). This subject role that causes them to accept or produce mismatched verbs with inanimate plural subject nouns that do not have a definite agent or patient role. If their feel free to accept or use mismatched verb. Experiments 1 and 2 w ere highly related, but Experiments 3, 4, and 5 test different factors. As I will address more below, the results in Chapters 3 and 4 clearly also signal the importance of semantic factors and are consistent with a constraint based, maximalist approach in which, in addition to syntactic information, other factors including semantic factors influence agreement during grammatical encoding and that agreement is not governed solely by syntactic information. Experiment 3 Experiment 3 tested the effect of unity v s. individuality of the entities of inanimate plural subject nouns in optional subject verb number agreement. The materials of this

PAGE 225

225 completion task included preambles with an inanimate subject head noun, a modifying prepositional phrase, and the nonverbal part of the compound verb. spatial separation vs. spatial collection I investigated the effect of collective and distributive conceptualization of the subject on the form of the produced verbs. The Individuality vs. Unity of the sub ject entities were manipulated by using different prepositions with the modifiers of the inanimate plural subjects. The preambles were one of two similar structures; passive or inchoative auxiliary. In order to test the effect of the plurality of the local noun, the number of the local noun were manipulated. So, there were three types of modifying prepositional phrase: Individualized Plural, Unified Plural, and Unified Singular. The results of this study showed that type of the prepositional phrase had a s ignificant effect on the production of mismatched verbs: participants completed the preambles with more mismatched verbs in sentences with unified prepositional phrases compared to individualized prepositional phrases. Comparing the proportion of the misma tched verbs produced in sentences with unified singular and unified plural prepositional phrases also showed a small attraction effect for local nouns. Similar to the results from Experiments 1 and 2, I believe that the conceptualization of subject head n oun is the crucial factor here. Grammatically, the subject head nouns of the sentences are plural nouns, but they can have matched (plural) or mismatched (singular) verbs. This experiment showed that the selection of the verb depends on whether the speaker s conceptualize the entities of the subject nouns as a unit or as individuals. If they conceptualize the subject nouns as a unit, they produce simple verbs, but if they conceptualize them as individuals then they use plural nouns.

PAGE 226

226 This result, similar to t he results of the previous experiments, supports the constraint based, maximalist approach, as both conceptual and syntactic information influence agreement and thus agreement in Persian is clearly not governed solely by syntactic information. Experiment 4 In Experiment 4, the effect of concreteness vs. abstractness of subject nouns in subject verb number agreement was tested. The preambles of this completion task consisted of an inanimate plural subject head noun and the non verbal part of the compound ver bs in passive and inchoative auxiliary structures. The results of this study showed that the concreteness vs. abstractness of the subject nouns was a significant factor in agreement, as participants produced significantly more mismatched verbs with abstra ct subject nouns compared to concrete ones. This result resembles the subject conceptualization results of Experiment 3: When the subjects of sentences are concrete entities, it is easier for speakers to conceptualize them as individual entities and not u se mismatched verbs for them. But, speakers use more mismatched verbs when the subjects of the sentences are abstract nouns. In this case, the speakers do not conceptualize the subject entities as multiple entities in fact the abstract nouns are not mult iple entities; we have arbitrarily used plural morphemes for them as we do for the concrete nouns. This is somehow similar to the grammatical gender (feminine or masculine) used for nouns in languages which have gender. I believe that abstract plural nouns are mostly conceptualized as a unit (their conceptual meaning) and not individualized entities (their grammatical form). However, in this for example using a ju dgment task to test whether participants consider the plural

PAGE 227

227 abstract nouns as a unit or individual. A future study would be helpful to confirm whether the conceptualization of abstract plural noun is like a unit or individual entities. Again, similar to the other experiments in this dissertation, the effect of abstractness vs. concreteness of subject nouns is the most consistent with the constraint based, maximalist approach that supports the idea that in addition to syntactic information, other factors ( including semantic factors) influence agreement. Experiment 5 In Experiment 5, the attraction of intermediary main object nouns in optional subject verb number agreement was tested. In this experiment, participants read preambles that consisted of an inani mate plural subject noun, an animate/inanimate singular/plural object noun, direct object marker, and the non verbal part of a compound verb. They then completed the sentences with a verbal part of the compound verbs. The effects of three features of the o bject were tested: number, animacy, and morphology (comparing the attraction effect of two different plural morphemes used to mark the animate plural objects nouns). The results of this experiment showed that the number feature was effective: participants produced more mismatched (singular) verbs when the intermediate object was singular and vice versa. While the animacy of the object noun was not influential, there was an effect of morphology: Participants produced fewer mismatched (singular) verbs when t he plural suffix of the intermediary animate plural object nouns was / an/ rather than / ha/. These results show that intermediary main object nouns may serve as attractors in optional subject verb number agreement. The first result, the attraction effect of the intermediary object noun (in SOV) sentence, is important because it shows an attraction effect of a local noun outside of the

PAGE 228

228 subject noun phrase and not in a subordinate clause. Previous studies have tested the attraction effect of local nouns mod ifier of the subject noun phrase (e.g. example 8.1, Bock & Miller 1991), the noun phrase of a relative clause post modifier of the subject phrase (example 8.2, Bock & Cutting 1992), or the cliticized object which is the intermediary object pronoun (exampl e 8.3, Franck, Soare, Frauenfelder, Rizzi 2009). (8.9) (8.10) (8.11) Le snateur les RECOIT The senator them RECEIVES On the other h and, Hartsuiker, Antn Mndez, & van Zee (2001) did tested the attraction effect of intermediary object noun in complement clauses in Dutch. No study has tested the attraction effect of main object nouns in subject verb agreement. The unmarked SOV word ord er of Persian language provided this opportunity to test the effect of an intermediary object. Also, the optionality of subject verb number agreement in Persian enabled me to test the attraction effect of main object nouns in grammatical Persian sentences (not using elicitation of errors or the ambiguous target sentences). The result for the effect of morphology here is very interesting, because the participants have produced different forms of the verbs to complete two preambles which are exactly the same (inanimate plural subject noun, animate plural object noun, direct object (accusative) marker, and non verbal part of the compound verb) except for the plural morpheme of the intermediary object. In one preamble the plural suffix was / an/

PAGE 229

229 used specificall y for animate nouns and in the other preamble, the plural suffix marker was / ha/ used for all animate or inanimate nouns. The results of Experiment 5 support the constraint based account, which is able to explain attraction effects in the same framework as it does for regular, error free agreement. The Constraint situation in which semantic and grammatical constraints apply within the normal range for agreement (that is, normal, error free agreement take s place because the cues align). But, variations like attraction happen if the semantic, structural, or morphological constraints are outside the normal domain. In the case of the attraction effects for object nouns seen in this dissertation (i.e. the effe ct of number and different plural morphemes of object), both syntactic and morphological factors of an element outside the scope of subject and verb have influence agreement during grammatical encoding and so this supports the idea that agreement is not go verned solely by syntactic information. Taken together, the results of the experiments of this dissertation showed effects verb type and the thematic role that verbs assign to their subject nouns, verb tense, conceptualization of entities of subject nouns as a unit or individuals concreteness/abstractness of the subject noun, and attraction for intervening main object nouns in optional subject verb number agreement. These factors all caused Persian native speakers to produce different proportions of misma tched but grammatical verbs to complete sentences. These results clearly establish the effect of both syntactic and non syntactic factors in the optional subject verb number agreement of Persian. More specifically, they show that non syntactic information of the verb, subject and even the intervening object

PAGE 230

230 can affect the number agreement. To reiterate, we saw that verb number, verb tense and the thematic role it assigns to the subject were influential, as was the conceptual information of the subject noun such as its unity and concreteness. The attraction effect of object noun (the effect of number feature of object noun) shows that subject verb agreement is not syntactically insulated: The number feature of a local noun outside the subject phrase can caus e Persian speakers to produce sentences with matched or mismatched verbs. Finally, the effect of morphology (use of specific plural morpheme with the intervening object) shows the effect of morphology factors of a noun phrase outside of the subject noun ph rase; the object noun with the object marker /ro/. Considering the exceptional characteristics of Persian mentioned at the beginning of this chapter including optional subject verb number agreement and having grammatical sentences as the data of the analys is, SOV word order, use of compound verbs, and different plural markers will increase the value of the results acquired by the experiments. The data of this dissertation are a good source for the models of language production and processing. Finally, I wan t to show how a constraint based model would specifically account for the results of the experiments of this dissertation on optional subject verb number agreement. In this framework, computing agreement is not the product of copying the number feature of the subject noun phrase onto the verb. It is a constraint based process in which multiple cues are integrated in the production of the inflected verb form. In addition to syntactic factors, type of the verb and the thematic role it assigns to its subject, verb tense, conceptualization of the entities of the subject noun (as a unit or as

PAGE 231

231 individuals), number and type of plural morphemes of the intervening object noun are among the effective constraints. Message is the origin of notional number. The number features of the subject and the verb are parallel, independent products of the same information at the Message. This means that both subject and verb get their number features from the semantic structure independently and the number feature of the verb is not copied from the subject. The source of the number feature on the verb and the number feature on the subject is the same but the subject and verb may inflect different number features as subject and v erb are affected by different constraints at Functional level. For example, the subject noun may get a plural number feature because the notional meaning of the entities in the message is ion from the same message depending on whether the entities of the subject noun are conceptualized as a unit or individuals (the effect of unity vs. individuality and concreteness vs. abstractness of subject nouns). Also, the number inflection of the verb may be singular or plural depending on the type of the verb. Crucially, there can be more than one factor pushing the verb inflection to be singular, for example if the verb assigns patient role (not agent role) to its subject nou n and the action has happe ned in the past tense rather than present tense. And finally, the number inflection of the verb may be singular or plural depending on the number of the intervening object nouns and the type of the plural morpheme it has. In this way, for subject verb num ber agreement in Persian employs both grammatical mechanisms and number semantics interact to create agreement. If the

PAGE 232

232 resulting number features in subject and verb are the same, then the product is a matched agreement. But, if the number inflection on the verb is different form the number feature of the subject noun (e.g. mismatched agreement) a bidirectional reconciliation mechanism (such as unification) will happen to align the different features. So, each of the above mentioned factors in the Functio nal level will exert some The g rammatical mechanism has the main role because subject verb agreement is obligatory for all subject nouns except for those that are inanimate and plural. By the same logi c, the a nimacy of the subject noun as a morphological factor can be considered to be the second important constraint because optional agreement is only for the inanimate (and not animate) subject nouns. The effect of the conceptualization of the subject se ems to be more important in abstract vs. concrete nou n s compar ed to unity vs. individuality. In Experiment 3 the unity vs. individuality of enti ties of subject nouns were high lighted using spatial collection vs. spatial distributivity. So, in normal sente nces, the effect of co nceptualization may not be that effective. The constraint coming from the verb type seems to be more significant than verb tense : the effect of verb tense was shown only in production (suggesting a preference to use more mismatched v erb s with the past tense compar ed to present tense) but not in acceptance judgements ( with a greater rate of reject ion of mismatched verbs in the present tense ) Conversely, t he effect of verb type was present in both production and judgement tasks The ef fect of verb type was significant ly important ; almost no mismatched verb was used with verbs of agency (although the subject were inanimate

PAGE 233

233 nouns) So the effect of verb type seems to be more important than the effect of animacy of the subject. For the att raction effect of main object noun s the effects of number feature and different plural morphemes of intermediary object nouns on agreement seem to be less than the effect of different factors on the involving elements of the agreements (subject and verb). Suggestions for Further S tudy An eye tracking study would be a good way to test the effects of the factors tested here in the comprehension of agreement. The results of eye tracking could be very interesting because the target sentences are grammatical sentences with matched and mismatched verbs, and would be helpful for psycholinguistic theories models of language processing, because of the specific characteristics of Persian. Experiment 2 showed the effect of verb tense in the production of agreement. I do past tense. The results in Experiment 1 (role interpretation) did not show this effect, as the participants did not conceive more patient role for the subjects of past tense sentences comparing to present ones. So, a judgment task, testing the effect of verb tense in the volitional aspects of the subject noun for doing the action is recommended Experiment 5 of this dissertation tested th e attraction effect of an intervening main object noun on optional subject verb agreement. The object noun was [+specific, + definite ] followed by an obligatory direct object marker /ra/. Considering the effect of the object marker, a completion task testi ng the attraction effect of [+specific, definite] main object noun (i.e., without the direct object marker) is recommended. This study may

PAGE 234

234 show a greater attraction effect of for the intervening object on subject verb agreement. Also, it may show the eff ect of animacy features in agreement. Reviewing the experiments and their results, the difference between the proportion of mismatched verbs produced with verbs of instruments (group D, 0.081) in Experiment 2 (table 4 6) and the proportion of mismatched verbs in different verb conditions (APN, 0.161 ISO, 0.275) in Experiment 5 (table 7 6) is interesting. The proportion of mismatched verbs produced with the intransitive verbs of instruments is much less than the proportion of mismatched verbs produced wi th the transitive verbs of instrument with different object condition. I do not have any reason for this difference. A study with transitive and intransitive verbs would be useful to compare the effects of verb transitivity on optional agreement.

PAGE 235

235 APP ENDIX A GLOSS JUDGMENT TASK Table A 1. List of the stimuli for Judgment experiment (condition: Past Singular) item condition VT S timuli gloss 1 past SG A tnab ha gere xord rope pl knot collide.PAST.3SG the ropes be came knotted. the ropes became knotted. 2 past SG A znjir ha pich xord chain pl twist collide.PAST.3SG the chains became twisted. the chains became twisted. 3 past SG A dr ha rng xord door pl paint collide.PAST.3SG the doors became painted. the doors became painted. 4 past SG A frsh ha xak xord rug pl dust collide.PAST.3SG the rugs became dusty. the rugs beca me dusty. 5 past SG A ketab xune ha ronq gereft library pl flourishing take.PAST.3SG the libraries flourished. the libraries flourished. 6 past SG A dftr ha baqi mund not ebook pl left remain.PAST.3SG the notebooks remained. the notebooks remained. 7 past SG A sqf ha foru rixt roof pl down collapse.PAST.3SG the roofs collapsed. the roofs collapsed. 8 past SG A gerdu ha qel xord walnut pl spinning collide.PAST.3SG the walnuts spun. the walnuts spun.

PAGE 236

236 Table A 1. Continued item condition VT S timuli gloss 9 past SG A bste ha brgsht xord package pl returning collide.PAST.3SG the packages returned. the packages returned. 10 past SG A kfsh ha sor xord shoe pl sliding collide.PAST.3SG the shoes slid. the shoes slid. 11 past SG A mive ha hdr rft fruit pl wasting go.PAST.3SG the fruits was wasted. the fruits were wasted. 12 past SG A brg ha tekan xord leaf pl swaying collide.PAST.3SG the lea ves swayed. the leaves swayed. 13 past SG A sng ha foru rft stone pl sinking go.PAST.3SG the stones sunk. the stones sunk. 14 past SG A taxte ha shenavr mund board pl f loating remain.PAST.3SG the boards was floating. the boards were floating. 15 past SG B lubia ha jvane zd bean pl sprout strike.PAST.3SG the beans sprouted. the beans sprouted. 16 past SG B mixk ha gol krd carnation flower do.PAST.3SG the carnations flowered. the carnations flowered.

PAGE 237

237 Table A 1. Continued item condition VT S timuli gloss 17 past SG B saqe ha brg Dad stem pl leaf give.PAST.3SG the stems budded. the stems budded. 18 past SG B dune ha roshd krd seed pl growing do.PAST.3SG the seeds grew. the seeds grew. 19 past SG B t ufan ha foru neshst storm pl subsiding sit.PAST.3SG the storms subsided. the storms subsided. 20 past SG B xmir ha vr amd dough pl raising come.PAST.3SG the dough raised. the doug h raised. 21 past SG B dst ha vrm krd hand pl swelling do.PAST.3SG the hands swelled. the hands swelled. 22 past SG B zxm ha jush xord sore pl scabbing over collide.PAST.3SG the sores scabbed over. the sores scabbed over. 23 past SG B pa ha tavl zd foot pl blistering strike.PAST.3SG the feet blistered. the feet blistered. 24 past SG B ahn ha zng zd iron pl rust strike.PAST.3SG the irons rusted. the irons rusted.

PAGE 238

238 Table A 1. Continued item condition VT S timuli gloss 25 past SG B shaxe ha shokufe krd branch pl blossom do.PAST.3SG t he branches blossomed. the branches blossomed. 26 past SG B jvane ha qonche krd twig pl bud do.PAST.3SG the twigs budded. the twigs budded. 27 past SG B xiyar ha kpk zd cucumber pl mold strike.PAST.3SG the cucumbers molded. the cucumbers molded. 28 past SG B gndom ha rishe krd wheat pl root do.PAST.3SG the wheat took root. the wheat took root. 29 past SG C lamp ha nur dad lamp pl light give.PAST.3SG the lamps gave light. the lamps gave light. 30 past SG C ruzname ha xesh xesh krd newspaper pl rustling do.PAST.3SG the ne wspapers rustled. the newspapers rustled. 31 past SG C ash ha qol zd porridge pl bubble strike.PAST.3SG the porridge bubbled. the porridge bubbled. 32 past SG C atshfeshan ha fv ran krd volcano pl gush do.PAST.3SG the volcanoes gushed. the volcanoes gushed.

PAGE 239

239 Table A 1. Continued item condition VT S timuli gloss 33 past SG C boshke ha nsht krd barrel pl ooze do.PAST.3SG the barrels oozed. the barrels oozed. 34 past SG C seke ha jering jering K rd coin pl jingle do.PAST.3SG the coins jingled. the coins jingled. 35 past SG C bad ha zuze K eshid wind pl moan pull.PAST.3SG the winds moaned. the winds moaned. 36 past SG C cheraq ha su su Z d light pl flicker strike.PAST.3SG the lights flickered. the lights flickered. 37 past SG C mte ha vez vez K rd drill pl buzz do.PAST.3SG the drills buzzed. the drills buzzed. 38 past SG C livan ha brq Z d glass pl gleaming strike.PAST.3SG the glasses gleamed. the glasses gleamed. 39 past SG C sim ha jrqe Z d cord pl spark strike.PAST.3SG the electrical cords sparked. the electrical cords sparked. 40 past SG C rre ha sr o se da K rd saw pl noise do.PAST.3SG the saws made noise. the saws made noise.

PAGE 240

240 Table A 1. Continued item condition VT S timuli G loss 41 past SG C pnjre ha tq tq K rd window pl rattling do.PAST. 3SG the windows rattled. the windows rattled. 42 past SG C durbin ha felash Z d camera pl flashing strike.PAST.3SG the cameras flashed. the cameras flashed. 43 past SG D ferestnde ha moxabere krd beacon pl transmitting do.PAST.3SG the beacons transmitted. the beacons transmitted. 44 past SG D durbin ha ks gereft camera pl photo take.PAST.3SG the ca meras photographed. the cameras photographed. 45 past SG D helikupter ha chrx zd helicopter pl rotating strike.PAST.3SG the helicopters rotated. the helicopters rotated. 46 past SG D qayeq ha frar krd boat pl escaping do.PAST.3SG the boats escaped. the boats escaped. 47 past SG D radiyo ha ettela dad radio pl information give.PAST.3SG the radios informed. the radios informed. 48 past SG D asansor ha bala rft elevator pl up go.PAST.3SG the elevators went up. the elevators went up.

PAGE 241

241 Table A 1. Continued item condition VT S timuli gloss 49 past SG D taksi ha dowr zd taxi pl round strike.PAST.3SG the taxies turned around. the taxies turned around. 50 past SG D mte ha surax krd drill pl hole do.PAST.3SG the drills made a h ole. the drills made a hole. 51 past SG D vanet bar ha qachaq krd truck pl smuggling do.PAST.3SG the trucks smuggled. the trucks smuggled. 52 past SG D tank ha hmle krd tank pl attack do.PAST.3SG the tanks attacked. the tanks attacked. 53 past SG D pnke ha xonk krd fan pl cool do.PAST.3SG the fans cooled. the fans cooled. 54 past SG D hva peima ha prvaz krd airplane pl flight do.PAST.3SG the airplanes flew. the airplanes flew. 55 past SG D bolndgoo ha elam krd loudspeaker pl announcing do.PAST.3SG the loudsp eakers announced. the loudspeakers announced. 56 past SG D radar ha shenasa'i krd radar pl identify do.PAST.3SG the radars identified. the radars identified.

PAGE 242

242 Table A 1. Continued item condit ion VT S timuli gloss 57 past SG E baqlava ha cheshmk zd baklava pl winking strike.PAST.3SG the baklavas winked. the baklavas winked. 58 past SG E qablame ha tqlob krd pot pl cheating do.PAST.3SG the pots cheated. the pots cheated. 59 past SG E medad ha xm krd pencil pl frowning do.PAST.3SG the pencils frowned. the pencils frowned. 60 past SG E jaru ha rah rft broom pl way go.PAST.3SG the broom walked. the broom walked. 61 past SG E divar ha sut zd wall pl whistling strike.PAST.3SG the walls whistled. the walls whistled 62 past SG E rusri ha lbxnd zd scarf pl smiling strike.PAST.3SG the scarves smiled. the scarves smiled. 63 past SG E tr ha gerye krd perfume pl crying do.PAST.3SG the perfumes cried. the perfumes cried. 64 past SG E lule ha vrzesh krd pipe pl exercising do.PAST.3SG the pipes exercised. the pipes exercised.

PAGE 243

243 Table A 1. Continued item condition VT S timuli gloss 65 past SG E xodkar ha zanu zd pen pl kneeling strike.PAST.3SG the pens kneeled. the pens kneeled. 66 past SG E bolndgu ha movafeqt krd loudspeaker pl agreeing do.PAS T.3SG the loudspeakers agreed. the loudspeakers agreed. 67 past SG E goje ha fekr krd tomato pl thinking do.PAST.3SG the tomatoes thought. the tomatoes thought. 68 past SG E noxod ha da'va krd pea pl quarreling do.PAST.3SG the peas quarreled. the peas quarreled. 69 past SG E kamputer ha ta'zim krd computer pl bowing do.PAST.3SG the computers bowed. the computers bowed. 70 past SG E pnjre ha jrmesh krd window pl exercising do.PAST.3SG the windows exercised. the windows exercised. 71 past SG F golabi ha xorde shod pea r pl eaten become.PAST.3SG the pears was eaten. the pears were eaten. 72 past SG F jurab ha pushide shod sock pl worn become.PAST.3SG the socks was worn. the socks were worn.

PAGE 244

244 Table A 1. Continued item condition VT S timuli gloss 73 past SG F dastan ha xunde shod story pl R ead become.PAST.3SG the stories was read. the stories were read. 74 past SG F dstmal ha x ride shod handkerchief pl bought become.PAST.3SG the handkerchiefs was bought. the handkerchiefs were bought. 75 past SG F damn ha duxte shod skirt pl sewn become.PAST.3SG the skirts was sew n. the skirts were sewn. 76 past SG F tpe ha D ide shod hill pl S een become.PAST.3SG the hills was seen. the hills were seen. 77 past SG F boshqab ha shoste shod plate pl was hed become.PAST.3SG the plates was washed. the plates were washed. 78 past SG F qndan ha shekste shod sugar bowl pl broken become.PAST.3SG the sugar bowls was broken. the sugar bowls were broken. 79 past SG F chmedan ha foruxte shod suitcase pl S old become.PAST.3SG the suitcases was sold. the suitcases were sold. 80 past SG F tut frngi ha chide shod strawbe rry pl picked become.PAST.3SG the strawberries was picked. the strawberries were picked.

PAGE 245

245 Table A 1. Continued item condition VT S timuli gloss 81 past SG F zhaket ha B afte shod sweater pl knitted become.P AST.3SG the sweaters was knitted. the sweaters were knitted. 82 past SG F eink ha shomorde shod eye glass pl counted become.PAST.3SG the eye glasses was counted. the eye glasses were counted 83 past SG F ngoshtr ha ferestade shod ring pl S ent become.PAST.3SG the rings was sent. the rings were sent. 84 past SG F hvij ha P oxte shod carrot pl cooked become.PAST.3SG the carrots was cooked. the carrots were cooked. 85 past SG G miz ha chubi bud table pl wooden be.PAST.3SG the tables was wooden. the tables were wooden. 86 past SG G meda d ha qermez bud pencil pl R ed be.PAST.3SG the pencils was red. the pencils were red. 87 past SG G sib ha S eft bud apple pl C risp be.PAST.3SG the apples was crisp. the apples were crisp 88 past SG G goje ha T aze bud tomato pl F resh be.PAST.3SG the tomatoes was fresh. the tomatoes were fresh.

PAGE 246

246 Table A 1. Continued item condition VT S timuli gloss 89 past SG G xane ha d ur bud house pl F ar be.PAST.3SG the houses was far. the houses were far. 90 past SG G nrde ha f elezi bud fence pl metallic be.PAST.3SG the fences was metallic. the fences were metall ic. 91 past SG G pele ha sngi bud stair pl made of stone be.PAST.3SG the stairs was made of stone. the stairs were made of stone. 92 past SG G sndli ha chrmi bud chair pl leath er be.PAST.3SG the chairs was leather. the chairs were leather. 93 past SG G qir ha s iyah bud tar pl b lack be.PAST.3SG the tar was black. the tar was black. 94 past SG G m drese ha nzdik bud school pl N ear be.PAST.3SG the schools was near. the schools were near. 95 past SG G txt ha bozorg bud bed pl B ig be.PAST.3SG the beds was big. the beds were big. 96 past SG G brf ha S efid bud snow pl W hite be.PAST.3SG the snow was white. the snow was white.

PAGE 247

247 Table A 1. Continued item condition VT stimuli gloss 97 past SG G gushvare ha qshn g bud ear ring pl beautiful be.PAST.3SG the ear rings was beautiful. the ear rings were beautiful. 98 past SG G shal ha brishmi bud shawl pl S ilk be.PAST.3SG the shawls was silk. the shawls were silk. 99 past SG H zmin ha sirab shod land pl saturated become.PAST.3SG the lands became saturated. the lands were saturated. 100 past SG H sa't ha gom shod watch pl los t become.PAST.3SG the watches became lost. the watches were lost. 101 past SG H lke ha pak shod stain pl wiped become.PAST.3SG the stains became wiped. the stain s were wiped. 102 past SG H sbzi ha pzhmorde shod herb pl withered/faded become.PAST.3SG the herbs became withered/faded. the herbs were withered/faded. 103 past SG H qayeq ha qrq shod boat pl sunk become.PAST.3SG the boats became sunk. the boats were sunk. 104 past SG H badkonak ha V el shod balloon pl floated away become.PAST.3SG the balloons became floated away. the balloons were floated away.

PAGE 248

248 Ta ble A 1. Continued item condition VT stimuli gloss 105 past SG H balon ha rha shod hot air balloon pl floated away become.PAST.3SG the hot air balloons became floated away. the hot air balloons were floated awa y. 106 past SG H nbar ha xrab shod warehouse pl ruined become.PAST.3SG the warehouses became ruined the warehouses were ruined 107 past SG H seke ha peida shod coin pl found become .PAST.3SG the coins became found. the coins were found. 108 past SG H siman ha sxt shod cement pl set become.PAST.3SG the cement became set. the cement was set. 109 past SG H bstni ha ab shod ice cream pl water become.PAST.3SG the ice cream became melted. the ice cream was melted. 110 past SG H rng ha xoshk S hod paint pl dry become.PAST.3SG the paint became dry. the paint dried. 111 past SG H varzeshgah ha nabud S hod stadium pl ruined become.PAST.3SG the stadiums became ruined. the stadiums were ruined. 112 past SG H toxm e morqha tmum S hod egg EZ hen pl end become.PAST.3SG the eggs became end. the eggs ran out.

PAGE 249

249 Table A 2. List of the stimuli for judgment experiment (Condition: Past Plural) item condition VT stimuli G loss 1 past PL A tnab ha gere xord n rope pl knot collide.Past 3PL the ropes became knotted. the ropes became knotted. 2 past PL A znjir ha pich xord n chain pl twist collide.Past 3PL the chains became twisted. the chains became twisted. 3 past PL A dr ha rng xord n door pl paint collide.Past 3PL the doors became painted. the doors became painted. 4 past PL A frsh ha xak xord n rug pl dust collide.Past 3PL the rugs became dusty. the rugs became dusty. 5 past PL A ketab xune ha ronq gereft n library pl flourishing take.Past 3PL the libraries flourished. the libraries flourished. 6 past PL A dftr ha baqi mund n notebook pl left remain.Past 3PL the notebooks remained. the notebooks remained. 7 past PL A sqf ha foru rixt n roof pl down collapse.Past 3PL the roofs collapsed. the roofs collapsed. 8 past PL A gerdu ha qel xord n walnut pl spinning collide.Past 3PL the walnuts spun. the walnuts spun.

PAGE 250

250 Table A 2. Continued item condition VT stimuli G loss 9 past PL A bste ha Brgsht xord n package pl returning collide.Past 3PL the packages returned. the packages returned. 10 past PL A k fsh ha sor xord n shoe pl sliding collide.Past 3PL the shoes slided. the shoes slided. 11 past PL A mive ha hdr rft n fruit pl wasting go.Past 3PL the fruits were wasted. the f ruits were wasted. 12 past PL A brg ha tekan xord n leaf pl swaying collide.Past 3PL the leaves swayed. the leaves swayed. 13 past PL A sng ha foru rft n stone pl sinking go.P ast 3PL the stones sunk. the stones sunk. 14 past PL A taxte ha shenavr mund n board pl floating remain.Past 3PL the boards were floating. the boards were floating. 15 past PL B lubia ha jvane zd n bean pl sprout strike.Past 3PL the beans sprouted. the beans sprouted. 16 past PL B mixk ha gol krd n C arnation flower do.Past 3PL the carnations flowered the carnations flowered.

PAGE 251

251 Table A 2. Continued item condition VT stimuli gloss 17 past PL B saqe ha brg dad n stem pl leaf give.Past 3PL the stems budded. the stems budded. 18 past PL B dune ha roshd krd n seed pl growing do.Past 3PL the seeds grew. the seeds grew. 19 past PL B tufan ha foru neshst n storm pl subsiding sit.Past 3PL the storms subsided. the storms subsided. 20 past PL B xmir ha vr amd n dough pl raising come.Past 3PL the dough raised. the dough raised. 21 past PL B dst ha vrm krd n hand pl swelling do .Past 3PL the hands swelled. the hands swelled. 22 past PL B zxm ha jush xord n sore pl scabbing over collide.Past 3PL the sores scabbed over. the sores scabbed over. 23 past PL B pa ha tavl zd n foot pl blistering strike.Past 3PL the feet blistered. the feet blistered. 24 past PL B ahn ha zng zd n iron pl rust strike.Past 3PL the irons rusted. the irons rusted.

PAGE 252

252 Table A 2. Continued item condition VT stimuli gloss 25 past PL B shaxe ha shokufe krd n branch pl blossom do.Past 3PL the branches blossomed. the branches blossomed. 26 past PL B jvane ha qonche krd n twig pl bud do.Past 3PL the twigs budded. the twigs budded. 27 past PL B xiyar ha kpk zd n cucumber pl mold strike.Past 3PL the cucumbers molded. the cucumbers molded. 28 past PL B gndom ha rishe krd n wheat pl root do.Past 3PL the wheat took root. the wheat took root. 29 past PL C lamp ha nur dad n lamp pl light give.Past 3PL the lamps gave light. the lamps gave light. 30 past PL C ruzname ha xesh xesh krd n newspaper pl rustling do.Past 3PL the newspapers rustled. the newspapers rustled. 3 1 past PL C ash ha qol zd n porridge pl bubble strike.Past 3PL the porridge bubbled. the porridge bubbled. 32 past PL C atshfeshan ha fvran krd n volcano pl gush do.Past 3PL the volcanoes gushed. the volcanoes gushed.

PAGE 253

253 Table A 2. Continued item condition VT stimuli gloss 33 past PL C boshke ha nsht krd n barrel pl ooze do.Past 3PL the barrels oozed. the bar rels oozed. 34 past PL C seke ha jering jering krd n coin pl jingle do.Past 3PL the coins jingled. the coins jingled. 35 past PL C bad ha zuze keshid n wind pl moan pull. Past 3PL the winds moaned. the winds moaned. 36 past PL C cheraq ha su su zd n light pl flicker strike.Past 3PL the lights flickered. the lights flickered. 37 past PL C mte ha vez vez krd n drill pl buzz do.Past 3PL the drills buzzed. the drills buzzed. 38 past PL C livan ha brq zd n glass pl gleaming strike.Past 3PL the glasses gleamed. the glasses gleamed. 39 past PL C sim ha jrqe zd n cord pl spark strike.Past 3PL the electrical cords sparked. the electrical cords sparked. 40 past PL C rre ha sr o seda krd n saw pl noise do.Past 3PL the saws made noise. the saws made noise.

PAGE 254

254 Table A 2. Continued item condition VT stimuli gloss 41 past PL C pnjre ha tq tq krd n window pl rattling do.Past 3PL the windows rattled. the windows rattled. 42 past PL C durbin ha felash zd n camera pl flashing strike.Past 3PL the cameras flashed. the cameras flashed. 43 past PL D fer estnde ha moxabere krd n beacon pl transmitting do.Past 3PL the beacons transmitted. the beacons transmitted. 44 past PL D durbin ha ks gereft n camera pl photo take.Past 3PL the ca meras photographed. the cameras photographed. 45 past PL D helikupter ha chrx zd n helicopter pl rotating strike.Past 3PL the helicopters rotated. the helicopters rotated. 46 past PL D qayeq ha frar krd n boat pl escaping do.Past 3PL the boats escaped. the boats escaped. 47 past PL D radiyo ha ettela dad n radio pl information give.Past 3PL the radios inf ormed. the radios informed. 48 past PL D asansor ha bala rft n elevator pl up go.Past 3PL the elevators went up. the elevators went up.

PAGE 255

255 Table A 2. Continued item condition VT stimuli gloss 4 9 past PL D taksi ha dowr zd n taxi pl round strike.Past 3PL the taxies turned around. the taxies turned around. 50 past PL D mte ha surax krd n drill pl hole do.Past 3PL the drills made a hole. the drills made a hole. 51 past PL D vanet bar ha qachaq krd n truck pl smuggling do.Past 3PL the trucks smuggled. the trucks smuggled. 52 past PL D tank ha hmle krd n tank pl attack do.Past 3PL the tanks attacked. the tanks attacked. 53 past PL D pnke ha xonk krd n fan pl cool do.Past 3PL the fans cooled. the fans c ooled. 54 past PL D hva peima ha prvaz krd n airplane pl flight do.Past 3PL the airplanes flew. the airplanes flew. 55 past PL D bolndgoo ha elam krd n loudspeaker pl announcing do.Past 3PL the loudspeakers announced. the loudspeakers announced. 56 past PL D radar ha shenasa'i krd n radar pl identify do.Past 3PL the radars identified. the radars ide ntified.

PAGE 256

256 Table A 2. Continued item condition VT stimuli gloss 57 past PL E baqlava ha cheshmk zd n baklava pl winking strike.Past 3PL the baklavas winked. the baklavas winked. 58 past PL E qablame ha tqlob krd n pot pl cheating do.Past 3PL the pots cheated. the pots cheated. 59 past PL E medad ha xm krd n pencil pl frowning do.Past 3PL the pencils frowned. the pencils frowned. 60 past PL E jaru ha rah rft n broom pl way go.Past 3PL the broom walked. the broom walked. 61 past PL E divar ha sut zd n wall pl whistling strike. Past 3PL the walls whistled. the walls whistled. 62 past PL E rusri ha lbxnd zd n scarf pl smiling strike.Past 3PL the scarves smiled. the scarves smiled. 63 past PL E tr ha gerye krd n perfume pl crying do.Past 3PL the perfumes cried. the perfumes cried. 64 past PL E lule ha vrzesh krd n pipe pl exercising do.Past 3PL the pipes exercised. the pipes exercised.

PAGE 257

257 Table A 2. Continued item condition VT stimuli gloss 65 past PL E xodkar ha zanu zd n pen pl kneeling strike.Past 3PL the pens kneeled. the pens kneeled. 66 past PL E bolndgu ha movafeqt krd n loudspeaker pl agreeing do.Past 3PL the loudspeakers agreed. the loudspeakers agreed. 67 past PL E goje ha fekr krd n tomato pl thinking do.Past 3PL the tomatoes thought. the tomatoes thought. 68 past PL E noxod ha da'va krd n pea pl quarreling do.Past 3PL the peas quarreled. the peas quarreled. 69 past PL E kamput er ha ta'zim krd n computer pl bowing do.Past 3PL the computers bowed. the computers bowed. 70 past PL E pnjre ha jrmesh krd n window pl exercising do.Past 3PL the windows exercised. the windows exercised. 71 past PL F golabi ha xorde shod n pear pl eaten become.Past 3PL the pears were eaten. the pears were eaten. 72 past PL F jurab ha pushide shod n sock pl worn become.Past 3PL the socks were worn. the socks were worn.

PAGE 258

258 Table A 2. Continued item condition VT stimuli G loss 73 past PL F dastan ha xande shod n story pl read become.Past 3PL the stories wer e read. the stories were read. 74 past PL F dstmal ha xride shod n handkerchief pl bought become.Past 3PL the handkerchiefs were bought. the handkerchiefs were bought. 75 past PL F damn ha duxte shod n skirt pl sewn become.Past 3PL the skirts were sewn. the skirts were sewn. 76 past PL F tpe ha dide shod n hill pl seen become.Past 3PL t he hills were seen. the hills were seen. 77 past PL F boshqab ha shoste shod n plate pl washed become.Past 3PL the plates were washed. the plates were washed. 78 past PL F qnda n ha shekste shod n sugar bowl pl broken become.Past 3PL the sugar bowls were broken. the sugar bowls were broken. 79 past PL F chmedan ha foruxte shod n suitcase pl sold become.Past 3PL the suitcases were sold. the suitcases were sold.

PAGE 259

259 Table A 2. Continued item condition VT stimuli gloss 80 past PL F tut frngi ha chide shod n strawberry pl picked become.Past 3PL the strawber ries were picked. the strawberries were picked. 81 past PL F zhaket ha bafte shod n sweater pl knitted become.Past 3PL the sweaters were knitted. the sweaters were knitted. 82 past PL F eink ha shomorde shod n eye glass pl counted become.Past 3PL the eye glasses were counted. the eye glasses were counted. 83 past PL F ngoshtr ha ferestade shod n ring pl sent become. Past 3PL the rings were sent. the rings were sent. 84 past PL F hvij ha poxte shod n carrot pl cooked become.Past 3PL the carrots were cooked. the carrots were cooked. 85 past PL G miz ha chubi bud n table pl wooden be.Past 3PL the tables were wooden. the tables were wooden. 86 past PL G medad ha qermez bud n pencil pl red be.Past 3PL the pencils were red the pencils were red. 87 past PL G sib ha seft bud n apple pl crisp be.Past 3PL the apples were crisp. the apples were crisp.

PAGE 260

260 Table A 2. Continued item condition VT stimuli gloss 88 past PL G goje ha taze bud n tomato pl fresh be.Past 3PL the tomatoes were fresh. the tomatoes were fresh. 89 past PL G xane ha dur bud n house pl far be.Past 3PL the houses were far the houses were far. 90 past PL G nrde ha felezi bud n fence pl metallic be.Past 3PL the fences were metallic. the fences were metallic. 91 past PL G pele ha sngi bud n stair pl made of stone be.Past 3PL the stairs were made of stone. the stairs were made of stone. 92 past PL G sndli ha chrmi bud n chair pl leather be.Past 3PL the chairs were leather. the chairs were leather. 93 past PL G qir ha siyah bud n tar pl black be.Past 3PL the tar were black. the tar was black. 94 past PL G mdrese ha nzdik bud n school pl near be.Past 3PL the schools were near. the schools were near. 95 past PL G txt ha bozorg bud n bed pl big be.Past 3PL the beds were big. the beds were big.

PAGE 261

261 Table A 2. Continued item conditi on VT stimuli gloss 96 past PL G brf ha sefid bud n snow pl white be.Past 3PL the snow were white. the snow was white. 97 past PL G gushvare ha qshng bud n ear ring pl beau tiful be.Past 3PL the ear rings were beautiful. the ear rings were beautiful. 98 past PL G shal ha brishmi bud n shawl pl silk be.Past 3PL the shawls were silk. the shawls were silk. 99 past PL H zmin ha sirab shod n land pl saturated become.Past 3PL the lands became saturated. the lands were saturated. 100 past PL H sa't ha gom shod n watch pl lost become.Past 3PL the watches became lost. the watches were lost. 101 past PL H lke ha pak shod n stain pl wiped become.Past 3PL the stains became wiped. the stains were wiped. 102 past PL H sbzi ha pzhmorde shod n herb pl withered/faded become.Past 3PL the herbs became withered/faded. the herbs were withered/faded. 103 past PL H qayeq ha qrq shod n boat pl sunk become.Past 3PL the boats became sunk. the boats were sunk.

PAGE 262

262 Table A 2. Continued item condition VT stimuli gloss 104 past PL H badkonak ha v el shod n balloon pl floated away become.Past 3PL the balloons became floated a way. the balloons were floated away. 105 past PL H balon ha r ha shod n hot air balloon pl floated away become.Past 3PL the hot air balloons became floated away. the hot air balloons were floated away. 106 past PL H nbar ha x rab shod n warehouse pl r uined become.Past 3PL the warehouses became ruined the warehouses were ruined 107 past PL H seke ha peida shod n coin pl found be come.Past 3PL the coins became found. the coins were found. 108 past PL H siman ha sxt shod n cement pl set become.Past 3PL the cement became set. the cement were set. 109 past PL H bstni ha ab shod n ice cream pl water become.Past 3PL the ice cream became melted. the ice cream was melted. 110 past PL H rng ha xoshk shod n paint pl dry become.Past 3PL the pain t became dry. the paint dried.

PAGE 263

263 Table A 2. Continued item condition VT stimuli gloss 111 past PL H varzeshgah ha nabud shod n stadium pl ruined become.Past 3PL the stadiums became ruined. t he stadiums were ruined. 112 past PL H toxm e morqha tmum shod n egg EZ hen pl end become.Past 3PL the eggs became end. the eggs ran out.

PAGE 264

264 Table A 3. List of the stimuli for judgment experiment (Condi tion: Present Singular) item conditon VT stimuli Gloss 1 Present SG A tnab ha gere mi xor e rope pl knot IND collide 3SG the ropes becomes knotted. the ropes become knotted. 2 Present SG A znjir ha pich mi xor e chain pl twist IND collide 3SG the chains becomes twisted. the chains become twisted. 3 Present SG A dr ha rng mi xor e door pl paint IND collide 3SG the d oors becomes painted. the doors become painted. 4 Present SG A frsh ha xak mi xor e rug pl dust IND collide 3SG the rugs becomes dusty. the rugs become dusty. 5 Present SG A ketab xune ha ronq mi gir e library pl flourishing IND take 3SG the libraries flourishes. the libraries flourish. 6 Present SG A dftr ha baqi mi mun e notebook pl left IND remain 3SG the notebooks remains. the notebooks remain. 7 Present SG A sqf ha foru mi riz e roof pl down IND collape 3SG the roofs collapses. the roofs collapse. 8 Present SG A gerd u ha qel mi xor e walnut pl spinning IND collide 3SG the walnuts spins. the walnuts spin.

PAGE 265

265 Table A 3. Continued item conditon VT stimuli Gloss 9 Present SG A bste ha brgsht mi xor e packa ge pl returning IND collide 3SG the packages returns. the packages return. 10 Present SG A kfsh ha sor mi xor e shoe pl sliding IND collide 3SG the shoes slid. the shoes slid. 11 Present SG A mive ha hdr mi r e fruit pl wasting IND go 3SG the fruits is wasted. the fruits are wasted. 12 Present SG A brg ha tekan mi xor e leaf pl swaying IND collide 3SG the leaves sways. the leaves sway. 13 Present SG A sng ha foru mi r e stone pl sinking IND go 3SG the stones sinks. the stones sink. 14 Present SG A taxte ha shenavr mi mu n e board pl floating IND remain 3SG the boards is floating. the boards are floating. 15 Present SG B lubia ha jvane mi zn e bean pl sprout IND strike 3SG the beans sprouts. the beans sprout. 16 Present SG B mixk ha gol mi kon e carnation flower IND DO 3SG the carnations flowers. the carnations flower.

PAGE 266

266 Table A 3. Continued item conditon VT stimuli Gloss 17 Present SG B saqe ha brg mi d e stem pl leaf IND give 3SG the stems buds. the stems bud. 18 Present SG B dune ha roshd mi kon e seed pl growing IND DO 3SG the seeds grows. the seeds grow. 19 Present SG B tufan ha foru mi shin e storm pl subsiding IND sit 3SG the storms subsides. the storms subside. 20 Present SG B xmir ha vr mi a d dough pl raising IND come 3SG the dough raises. the dough raises. 21 Present SG B dst ha vrm mi kon e hand pl swelling IND DO 3SG the hands swells. the hands swell. 22 Present SG B zxm ha jush mi xor e sore pl scabbing over IND collide 3SG the sores scabs over. the sores scab over. 23 Present SG B pa ha tavl mi zn e foot pl blistering IND strike 3SG the feet blisters. the feet blister. 24 Present SG B ahn ha zng mi zn e iron pl rust IND strike 3SG the irons rusts. the irons rust.

PAGE 267

267 Table A 3. Continued item conditon VT stimuli Gloss 25 Presen t SG B shaxe ha shokufe mi kon e branch pl blossom IND DO 3SG the branches blossoms. the branches blossom. 26 Present SG B jvane ha qonche mi kon e twig pl bud IND DO 3 SG the twigs buds. the twigs bud. 27 Present SG B xiyar ha kpk mi zn e cucumber pl mold IND strike 3SG the cucumbers molds. the cucumbers mold. 28 Present SG B g ndom ha rishe mi kon e wheat pl root IND DO 3SG the wheat takes root. the wheat takes root. 29 Present SG C lamp ha nur mi d e lamp pl light IND give 3SG the lamps gives light. the l amps give light. 30 Present SG C ruzname ha xesh mi kon e newspaper pl rustling IND DO 3SG the newspapers rustles. the newspapers rustle. 31 Present SG C ash ha qol mi zn e porridge pl bubble IND strike 3SG the porridge bubbles. the porridge bubbles. 32 Present SG C atshfeshan ha fvran mi kon e volcano pl gush IND DO 3SG the volcanoes gushes. th e volcanoes gush.

PAGE 268

268 Table A 3. Continued item conditon VT stimuli Gloss 33 Present SG C boshke ha nsht mi kon e barrel pl ooze IND DO 3SG the barrels oozes. the barrels ooze. 34 Present SG C seke ha jering jering mi kon e coin pl jingle IND DO 3SG the coins jingles. the coins jingle. 35 Present SG C bad ha zuze mi kesh e wind pl moan IND pull 3SG the winds m oans. the winds moan. 36 Present SG C cheraq ha su su mi zn e light pl flicker IND strike 3SG the lights flickers. the lights flicker. 37 Present SG C mte ha vez vez mi kon e drill pl buzz IND DO 3SG the drills buzzes. the drills buzz. 38 Present SG C livan ha brq mi zn e glass pl gleaming IND strike 3SG the glasses gleams. the glasses gleam. 39 Present SG C sim ha jrqe mi zn e cord pl spark IND strike 3SG the electrical cords sparks. the electrical cords spark. 40 Present SG C rre ha sr o seda mi kon e s aw pl noise IND DO 3SG the saws makes noise. the saws make noise.

PAGE 269

269 Table A 3. Continued item conditon VT stimuli Gloss 41 Present SG C pnjre ha tq tq mi kon e window pl rattling IND DO 3SG the windows rattles. the windows rattle. 42 Present SG C durbin ha felash mi zn e camera pl flashing IND strike 3SG the cameras flashes. the cameras flash. 43 Present SG D ferestnde ha moxabere mi kon e beacon pl transmitting IND DO 3SG the beacons transmits. the beacons transmit. 44 Present SG D durbin ha ks mi gir e camera pl photo IND take 3SG the cameras photographs. the cameras photograph. 45 Present SG D helikupter ha chrx mi zn e helicopter pl rotating IND strike 3SG the helicopters rotates. the helicopters rotate. 46 P resent SG D qayeq ha frar mi kon e boat pl escaping IND DO 3SG the boats escapes. the boats escape. 47 Present SG D radiyo ha ettela mi d e radio pl information IND give 3 SG the radios informs. the radios inform. 48 Present SG D asansor ha bala mi r e elevator pl up IND go 3SG the elevators goes up. the elevators go up.

PAGE 270

270 Table A 3. Continued item condi ton VT stimuli Gloss 49 Present SG D taksi ha dowr mi zn e taxi pl round IND strike 3SG the taxies turns around. the taxies turn around. 50 Present SG D mte ha surax mi kon e drill pl hole IND DO 3SG the drills makes a hole. the drills make a hole. 51 Present SG D vanet bar ha qachaq mi kon e truck pl smuggling IND DO 3SG the trucks smuggles. th e trucks smuggle. 52 Present SG D tank ha hmle mi kon e tank pl attack IND DO 3SG the tanks attacks. the tanks attack. 53 Present SG D pnke ha xonk mi kon e fan pl coo l IND DO 3SG the fans cools. the fans cool. 54 Present SG D hva peima ha prvaz mi kon e airplane pl flight IND DO 3SG the airplanes flies. the airplanes fly. 55 Present SG D bolndgoo ha elam mi kon e loudspeaker pl announcing IND DO 3SG the loudspeakers announces. the loudspeakers announce. 56 Present SG D radar ha shenasa'i mi kon e radar pl identify IND DO 3SG the radars identifies. the radars identify.

PAGE 271

271 Table A 3. Continued item conditon VT stimuli Gloss 57 Present SG E baqlava ha cheshmk mi zn e baklava pl winking IND strike 3SG the baklavas winks. the baklavas wink. 58 Present SG E qablame ha tqlob mi kon e pot pl cheating IND DO 3SG the pots cheats. the pots cheat. 59 Present SG E m edad ha xm mi kon e pencil pl frowning IND DO 3SG the pencils frowns. the pencils frown. 60 Present SG E jaru ha rah mi r e broom pl way IND go 3SG the broom walks. the broom walk. 61 Present SG E divar ha sut mi zn e wall pl whistling IND strike 3SG the walls whistles. the walls whistle. 62 Present SG E rusri ha lbxnd mi zn e scarf pl smiling I ND strike 3SG the scarves smiles. the scarves smile. 63 Present SG E tr ha gerye mi kon e perfume pl crying IND DO 3SG the perfumes cries. the perfumes cry. 64 Present SG E lule ha vrzesh mi kon e pipe pl exercising IND DO 3SG the pipes exercises. the pipes exercise.

PAGE 272

272 Table A 3. Continued item conditon VT stimuli Gloss 65 Present SG E xodkar ha zanu mi zn e pen pl kneeling IND strike 3SG the pens kneels. the pens kneel. 66 Present SG E bolndgu ha movafeqt mi kon e loudspeaker pl agreeing IND DO 3SG the loudspeakers agrees. the loudspeakers agree. 67 Present SG E goje ha fekr mi kon e tomato pl thinking IND DO 3SG the tomatoes thinks. the tomatoes think. 68 Present SG E noxod ha da'va mi kon e pea pl quarreling IND DO 3SG the peas quarrels. the peas quarrel. 69 Present SG E kamputer ha ta'zim mi kon e computer pl bowing IND DO 3SG the computers bows. the computers bow. 70 Present SG E pnjre ha jrmesh mi kon e window pl exercising IND DO 3SG the windows exercises. the windows exercise. 71 Present SG F golabi ha xorde mi sh e pear pl ea ten IND become 3SG the pears is eaten. the pears are eaten. 72 Present SG F jurab ha pushide mi sh e sock pl worn IND become 3SG the socks is worn. the socks are worn.

PAGE 273

273 Table A 3. Continued item conditon VT stimuli Gloss 73 Present SG F dastan ha xande mi sh e story pl read IND become 3SG the stories is read. the stories are read. 74 Present SG F dstm al ha xride mi sh e handkerchief pl bought IND become 3SG the handkerchiefs is bought. the handkerchiefs are bought. 75 Present SG F damn ha duxte mi sh e skirt pl sewn IND become 3SG th e skirts is sewn. the skirts are sewn. 76 Present SG F tpe ha dide mi sh e hill pl seen IND become 3SG the hills is seen. the hills are seen. 77 Present SG F boshqab ha shoste mi sh e plate pl washed IND become 3SG the plates is washed. the plates are washed. 78 Present SG F qndan ha shekste mi sh e sugar bowl pl broken IND become 3SG the sugar bowls is broke n. the sugar bowls are broken. 79 Present SG F chmedan ha foruxte mi sh e suitcase pl sold IND become 3SG the suitcases is sold. the suitcases are sold. 80 Present SG F tut frngi ha chide mi sh e strawberry pl picked IND become 3SG the strawberries is picked. the strawberries are picked.

PAGE 274

274 Table A 3. Continued item conditon VT stimuli Gloss 81 Present SG F zhaket ha bafte mi sh e sweater pl knitted IND become 3SG the sweaters is knitted. the sweaters are knitted. 82 Present SG F eink ha shomorde mi sh e eye glass pl counted IND become 3SG the eye glasses is counted. the eye glasses are counted. 83 Present SG F ngoshtr ha ferestade mi sh e ring pl sent IND become 3SG the rings is sent. the rings are sent. 84 Present SG F hvij ha poxte mi sh e carrot pl cooked IND become 3SG the carrots is cooked. the carrots are cooked. 85 Present SG G miz ha chubi ye table pl wooden be PRES 3SG the tables is wooden. the tables are wooden. 86 Present SG G medad ha qermez e pencil pl red be PRES 3SG the pencils is red. the pencils are red. 87 Present SG G sib ha seft e apple pl crisp be PRES 3SG the apples is crisp. the apples are crisp. 88 Present SG G goje ha taze s tomato pl fresh be PRES 3SG the tomatoes is fresh. the tomatoes are fresh.

PAGE 275

275 Table A 3. Continued item cond iton VT stimuli Gloss 89 Present SG G xane ha dur e house pl far be PRES 3SG the houses is far. the houses are far. 90 Present SG G nrde ha felezi ye fence pl metallic be PRES 3SG the fences is metallic. the fences are metallic. 91 Present SG G pele ha sngi ye stair pl made of stone be PRES 3SG the stairs is made of stone. the stairs are made of stone. 92 Present SG G sndli ha chrmi ye chair pl leather be PRES 3SG the chairs is leather. the chairs are leather. 93 Present SG G qir ha siyah e tar pl black be PRES 3SG the tar is black. the tar is black. 94 Present SG G mdrese ha nzdik e school pl near be PRES 3SG the schools is near. the schools are near. 95 Present SG G txt ha bozorg e bed pl big be PR ES 3SG the beds is big. the beds are big. 96 Present SG G brf ha sefid e snow pl white be PRES 3SG the snows is white. the snow is white.

PAGE 276

276 Table A 3. Continued item conditon VT stimuli Gl oss 97 Present SG G gushvare ha qshng e ear ring pl beautiful be PRES 3SG the ear rings is beautiful. the ear rings are beautiful. 98 Present SG G shal ha brishmi ye shawl pl silk be PRES 3SG the shawls is silk. the shawls are silk. 99 Present SG H zmin ha sirab mi sh e land pl saturated IND become 3SG the lands becomes saturated. the lands are saturated. 100 Present SG H sa't ha gom mi sh e watch pl lost IND become 3SG the watches becomes lost. the watches are lost. 101 Present SG H lke ha pak mi sh e stain pl wiped IND become 3SG the stains becomes wiped. the stains are wiped. 102 Present SG H sbzi ha pzhmorde mi sh e herb pl withered/faded IND become 3SG the herbs become s withered/faded. the herbs are withered/faded. 103 Present SG H qayeq ha qrq mi sh e boat pl sunk IND become 3SG the boats becomes sunk. the boats are sunk. 104 Present SG H badkonak ha vel mi sh e balloon pl floated away IND become 3SG the balloons becomes floated away. the balloons are floated away.

PAGE 277

277 Table A 3. Continued item conditon VT stimuli Gloss 105 Present SG H balon ha rha mi sh e hot air balloon pl floated away I ND become 3SG the hot air balloons becomes floated away. the hot air balloons are floated away. 106 Present SG H nbar ha xrab mi sh e warehouse pl ruined IND become 3SG the warehouses becomes ruine d the warehouses are ruined 107 Present SG H seke ha peida mi sh e coin pl found IND become 3SG the coins becomes found. the coins are found. 108 Present SG H siman ha sxt mi s h e cement pl set IND become 3SG the cement becomes set. the cement is set. 109 Present SG H bstni ha ab mi sh e ice cream pl water IND become 3SG the ice cream becomes melted. the ice cream is melted. 110 Present SG H rng ha xoshk mi sh e paint pl dry IND become 3SG the paint becomes dry. the paint dries. 111 Present SG H varzeshgah ha nabud mi sh e stadiu m pl ruined IND become 3SG the stadiums becomes ruined. the stadiums are ruined. 112 Present SG H toxm e morq ha tmum mi sh e egg EZ hen pl end IND become 3SG the eggs becomes end. the eg gs run out.

PAGE 278

278 Table A 4. List of the stimuli for judgment experiment (Condition: Present Plural) item condition VT stimuli gloss 1 Present PL A tnab ha gere mi xor n rope pl knot IND collide 3PL the ropes bec ome knotted. the ropes become knotted. 2 Present PL A znjir ha pich mi xor n the chains become twisted. the chains become twisted. 3 Present PL A dr ha rng mi xor n door pl paint IND collide 3PL the doors become painted. the doors become painted. 4 Present PL A frsh ha xak mi xor n rug pl dust IND collide 3PL the rugs become dusty. the rugs be come dusty. 5 Present PL A ketab xune ha ronq mi gir n library pl flourishing IND take 3PL the libraries flourish. the libraries flourish. 6 Present PL A dftr ha baqi mi mu n n notebook pl left IND remain 3PL the notebooks remain. the notebooks remain. 7 Present PL A sqf ha foru mi riz n roof pl down IND collapse 3PL the roofs collapse. the roofs co llapse. 8 Present PL A gerdu ha qel mi xor n walnut pl spinning IND collide 3PL the walnuts spin. the walnuts spin.

PAGE 279

279 Table A 4. Continued item condition VT stimuli gloss 9 Present PL A bste ha brgsht mi xor n package pl returning IND collide 3PL the packages return. the packages return. 10 Present PL A kfsh ha sor mi xor n shoe pl sliding IND collide 3PL th e shoes slid. the shoes slid. 11 Present PL A mive ha hdr mi r n fruit pl wasting IND go 3PL the fruits are waste. the fruits are waste. 12 Present PL A brg ha tekan mi x or n leaf pl swaying IND collide 3PL the leaves sway. the leaves sway. 13 Present PL A sng ha foru mi r n stone pl sinking IND go 3PL the stones sink. the stones sink. 14 Present PL A taxte ha shenavr mi mun n board pl floating IND remain 3PL the boards are floating. the boards are floating. 15 Present PL B lubia ha jvane mi zn n bean pl spro ut IND strike 3PL the beans sprout. the beans sprout. 16 Present PL B mixk ha gol mi kon n carnation flower IND DO 3PL the carnations flower. the carnations flower.

PAGE 280

280 Table A 4. Continu ed item condition VT stimuli gloss 17 Present PL B saqe ha brg mi d n stem pl leaf IND give 3PL the stems buds. the stems bud. 18 Present PL B dune ha roshd mi kon n seed pl growing IND DO 3PL the seeds grow. the seeds grow. 19 Present PL B tufan ha foru mi shin n storm pl subsiding IND sit 3PL the storms subside. the storms subside. 20 Present PL B xmir ha vr mi a n dough pl raising IND come 3PL the dough raise. the dough raises. 21 Present PL B dst ha vrm mi kon n hand pl swelling IND DO 3PL the hands swell. the hands swell. 22 Present PL B zxm ha jush mi xor n sore pl scabbing over IND collide 3PL the sores scab over. the sores scab over. 23 Present PL B pa ha tavl mi zn n foot pl blistering IND strike 3PL the feet blister. the feet blister. 24 Present PL B ahn ha zng mi zn n iron pl rust IND strike 3PL the irons rust. the irons rust.

PAGE 281

281 Table A 4. Continu ed item condition VT stimuli gloss 25 Present PL B shaxe ha shokufe mi kon n branch pl blossom IND DO 3PL the branches blossom. the branches blossom. 26 Present PL B jvane ha qonc he mi kon n twig pl bud IND DO 3PL the twigs bud. the twigs bud. 27 Present PL B xiyar ha kpk mi zn n cucumber pl mold IND strike 3PL the cucumbers mold. the cucumbers mold. 2 8 Present PL B gndom ha rishe mi kon n wheat pl root IND DO 3PL the wheat take root. the wheat takes root. 29 Present PL C lamp ha nur mi d n lamp pl light IND give 3PL the lamps give light. the lamps give light. 30 Present PL C ruzname ha xesh xesh mi kon n newspaper pl rustling IND DO 3PL the newspapers rustle. the newspapers rustle. 31 Present PL C ash ha qol mi zn n porridge pl bubble IND strike 3PL the porridge bubble. the porridge bubbles. 32 Present PL C atshfeshan ha fvran mi kon n volcano pl gush IND DO 3PL the volcanoes gush. the volcanoes gush.

PAGE 282

282 Table A 4. Continued item condition VT stimuli gloss 33 Present PL C boshke ha nsht mi kon n barrel pl ooze IND DO 3PL the barrels ooze. the barre ls ooze. 34 Present PL C seke ha jering jering mi kon n coin pl jingle IND DO 3PL the coins jingle. the coins jingle. 35 Present PL C bad ha zuze mi kesh n wind pl moan IND pull 3PL the winds moan. the winds moan. 36 Present PL C cheraq ha su su mi zn n light pl flicker IND strike 3PL the lights flicker. the lights flicker. 37 Present PL C mte ha vez vez mi kon n drill pl buzz IND DO 3PL the drills buzz. the drills buzz. 38 Present PL C livan ha brq mi zn n glass pl gleaming IND strike 3PL the glasses gleam. the glasses gleam. 39 Present PL C sim ha jrqe mi zn n cord pl spark IND strike 3PL the electrical cords spark. the electrical cords spark. 40 Present PL C rre ha sr o seda mi kon n saw pl noise IND DO 3PL the saws make noise. the saws make noise.

PAGE 283

283 Table A 4. Continued item condition VT stimuli gloss 41 Present PL C pnjre ha tq tq mi kon n window pl rattling IND DO 3PL the windows rattle. the windows rattle. 42 Present PL C durbin ha felash mi zn n camera pl flashing IND strike 3PL the cameras flash. the cameras flash. 43 Present PL D ferestnde ha moxabere mi kon n beacon pl transmitting IND DO 3PL the beacons transmit. the beacons transmit. 44 Present PL D durbin ha ks mi gir n camera pl photo IND take 3PL the cameras photograph. the cameras photograph. 45 Present PL D helikupter ha chrx mi zn n helicopter pl rotating IND strike 3PL the helicopters rotate. the helicop ters rotate. 46 Present PL D qayeq ha frar mi kon n boat pl escaping IND DO 3PL the boats escape. the boats escape. 47 Present PL D radiyo ha ettela mi d n radio pl infor mation IND give 3PL the radios inform. the radios inform. 48 Present PL D asansor ha bala mi r n elevator pl up IND go 3PL the elevators go up. the elevators go up.

PAGE 284

284 Table A 4. Cont inued item condition VT stimuli gloss 49 Present PL D taksi ha dowr mi zn n taxi pl round IND strike 3PL the taxies turn around. the taxies turn around. 50 Present PL D mte ha surax mi kon n drill pl hole IND DO 3PL the drills make a hole. the drills make a hole. 51 Present PL D vanet bar ha qachaq mi kon n truck pl smuggling IND DO 3PL the trucks smuggle. the trucks smuggle. 52 Present PL D tank ha hmle mi kon n tank pl attack IND DO 3PL the tanks attack. the tanks attack. 53 Present PL D pnke ha xonk mi kon n fan pl cool IND DO 3PL the fans cool. the fans cool. 54 Present PL D hva peima ha prvaz mi kon n airplane pl flight IND DO 3PL the airplanes fly. the airplanes fly. 55 Present PL D bolndgoo ha elam mi kon n loudspeaker pl announcing IND DO 3PL the loudspeakers announce. the loudspeakers announce. 56 Present PL D radar ha shenasa'i mi kon n radar pl ide ntify IND DO 3PL the radars identify. the radars identify.

PAGE 285

285 Table A 4. Continued item condition VT stimuli gloss 57 Present PL E baqlava ha cheshmk mi zn n baklava pl winking IND strike 3PL the baklavas wink. the baklavas wink. 58 Present PL E qablame ha tqlob mi kon n pot pl cheating IND DO 3PL the pots cheat. the pots cheat. 59 Present PL E medad ha x m mi kon n pencil pl frowning IND DO 3PL the pencils frown. the pencils frown. 60 Present PL E jaru ha rah mi r n broom pl way IND go 3PL the broom walk. the broom walk. 61 Presen t PL E divar ha sut mi zn n wall pl whistling IND strike 3PL the walls whistle. the walls whistle. 62 Present PL E rusri ha lbxnd mi zn n scarf pl smiling IND strike 3PL the scarves smile. the scarves smile. 63 Present PL E tr ha gerye mi kon n perfume pl crying IND DO 3PL the perfumes cry. the perfumes cry. 64 Present PL E lule ha vrzesh mi kon n pipe pl exercising IND DO 3PL the pipes exercise. the pipes exercise.

PAGE 286

286 Table A 4. Continued item condition VT stimuli gloss 65 Present PL E xodkar ha zanu mi zn n pen pl kneeling IND strike 3PL the pens kneel. the pens kneel. 66 Present PL E bolndgu ha movafeqt mi kon n loudspeaker pl agreeing IND DO 3PL the loudspeakers agree. the loudspeakers a gree. 67 Present PL E goje ha fekr mi kon n tomato pl thinking IND DO 3PL the tomatoes think. the tomatoes think. 68 Present PL E noxod ha da'va mi kon n pea pl quarreling IND DO 3PL the peas quarrel. the peas quarrel. 69 Present PL E kamputer ha ta'zim mi kon n computer pl bowing IND DO 3PL the computers bow. the computers bow. 70 Present PL E pnjre ha jrmesh mi kon n window pl exercising IND DO 3PL the windows exercise. the windows exercise. 71 Present PL F golabi ha xorde mi sh n pear pl eaten IND become 3PL the pears are eaten. the pears are eaten. 72 Present PL F jurab ha pushide mi sh n sock pl worn IND become 3PL the socks are worn. the socks are worn.

PAGE 287

287 Table A 4. Continued item co ndition VT stimuli gloss 73 Present PL F dastan ha xande mi sh n story pl read IND become 3PL the stories are read. the stories are read. 74 Present PL F dstmal ha xride mi s h n handkerchief pl bought IND become 3PL the handkerchiefs are bought. the handkerchiefs are bought. 75 Present PL F damn ha duxte mi sh n skirt pl sewn IND become 3PL the skirts are sewn. the skirts are sewn. 76 Present PL F tpe ha dide mi sh n hill pl seen IND become 3PL the hills are seen. the hills are seen. 77 Present PL F boshqab ha shoste mi sh n plate pl washed IND become 3PL the plates are washed. the plates are washed. 78 Present PL F qndan ha shekste mi sh n sugar bowl pl broken IND become 3PL the sugar bowls are broken. the sugar bowls are broken. 79 Present PL F chmedan ha foruxte mi sh n suitcase pl sold IND become 3PL the suitcases are sold. the suitcases are sold. 80 Present PL F tut frngi ha chide mi sh n strawberry pl picked IND become 3PL the strawberries are picked. the strawberries are picked.

PAGE 288

288 Table A 4. Continued item condition VT stimuli gloss 81 Present PL F zha ket ha bafte mi sh n sweater pl knitted IND become 3PL the sweaters are knitted. the sweaters are knitted. 82 Present PL F eink ha shomorde mi sh n eye glass pl counted IND become 3PL the eye glasses are counted. the eye glasses are counted. 83 Present PL F ngoshtr ha ferestade mi sh n ring pl sent IND become 3PL the rings are sent. the rings are sent. 84 Present PL F hvij ha poxte mi sh n carrot pl cooked IND become 3PL the carrots are cooked. the carrots are cooked. 85 Present PL G miz ha chubi n table pl wooden be PRES 3PL the tab les are wooden. the tables are wooden. 86 Present PL G medad ha qermez n pencil pl red be PRES 3PL the pencils are red. the pencils are red. 87 Present PL G sib ha seft n a pple pl crisp be PRES 3PL the apples are crisp. the apples are crisp. 88 Present PL G goje ha taze n tomato pl fresh be PRES 3PL the tomatoes are fresh. the tomatoes are fresh.

PAGE 289

289 Table A 4. Continued item condition VT stimuli gloss 89 Present PL G xane ha dur n house pl far be PRES 3PL the houses are far. the houses are far. 90 Present PL G nrde ha felezi n fence pl metallic be PRES 3PL the fences are metallic. the fences are metallic. 91 Present PL G pele ha sngi n stair pl made of stone be PRES 3PL the stairs are made of stone. the stairs are ma de of stone. 92 Present PL G sndli ha chrmi n chair pl leather be PRES 3PL the chairs are leather. the chairs are leather. 93 Present PL G qir ha siyah n tar pl black be PRES 3P L the tar are black. the tar is black. 94 Present PL G mdrese ha nzdik n school pl near be PRES 3PL the schools are near. the schools are near. 95 Present PL G txt ha bo zorg n bed pl big be PRES 3PL the beds are big. the beds are big. 96 Present PL G brf ha sefid n snow pl white be PRES 3PL the snow are white. the snow is white.

PAGE 290

290 Table A 4. Con tinued item condition VT stimuli gloss 97 Present PL G gushvare ha qshng n ear ring pl beautiful be PRES 3PL the ear rings are beautiful. the ear rings are beautiful. 98 Present PL G shal ha brishmi n shawl pl silk be PRES 3PL the shawls are silk. the shawls are silk. 99 Present PL H zmin ha sirab mi sh n land pl saturated IND become 3PL the lands become saturated. the lands are saturated. 100 Present PL H sa't ha gom mi sh n watch pl lost IND become 3PL the watches become lost. the watches are lost. 101 Present PL H lke ha pak mi sh n stain pl wiped IND become 3PL the stains become wiped. the stains are wiped. 102 Present PL H sbzi ha pzhmorde mi sh n vegetable pl withered/faded IND become 3PL the herbs become withered /faded. the herbs are withered/faded. 103 Present PL H qayeq ha qrq mi sh n boat pl sunk IND become 3PL the boats become sunk. the boats are sunk. 104 Present PL H badkonak ha vel mi sh n balloon pl floated away IND become 3PL the balloons become floated away. the balloons are floated away.

PAGE 291

291 Table A 4. Continued item condition VT stimuli gloss 105 Present PL H balon ha rha m i sh n balloon pl floated away IND become 3PL the hot air balloons become floated away. the hot air balloons are floated away. 106 Present PL H nbar ha xrab mi sh n warehouse pl ruined IND becom e 3PL the warehouses become ruined the warehouses are ruined 107 Present PL H seke ha peida mi sh n coin pl found IND become 3PL the coins become found. the coins are found. 108 Present PL H siman ha sxt mi sh n cement pl set IND become 3PL the cement become set. the cement becomes set. 109 Present PL H bstni ha ab mi sh n ice cream pl water IND become 3PL the ice cream become melted. the ice cream is melted. 110 Present PL H rng ha xoshk mi sh n paint pl dry IND become 3PL the paint become dry. the paint dries. 111 Present PL H varzeshgah ha nabud mi sh n stadium pl ruined IND become 3PL the stadiums become ruined. the stadiums are ruined. 112 Present PL H toxm e morq ha tmam mi sh n egg EZ hen pl end IND becom e 3PL the eggs become end. the eggs run out.

PAGE 292

292 Table A 5. List of the filler stimuli for judgme nt experiment (Condition: past) item stimuli gloss 1 mn v friborz msa'el e riyazi ro hl krd im I and Fariborz problems EZ math OM solving did 1PL Fariborz and I solved math problems. 2 mn v ehsan hva ye srd ro dust dasht im I and Ehsan weather EZ cold OM loving dad 1PL Ehsan and I liked cold weather. 3 shoma v mhshid emtehan dad in you and Mahshid exam gave 2PL Mahshid and you took the exam. 4 shoma v hsn ab e mahi ha ro vz krd in you and Hasan water EZ fish pl OM changing did 2PL Hasan and yo u changed the water of the fish bowl. 5 sara v mn z muze didn krd im Sara and I from museum visiting did 1PL Sara and I visited the mu seum. 6 shhab v shoma drs ha ro morur krd in Shahab and you lesson pl OM reviewing did 2PL Shahab and you reviewed the lessons. 7 kaveh v mn hmkelas bud im Kaveh and I classmate were 1PL Kaveh and I were classmates. 8 nstrn v shoma gol ha ro ab dad in Nastaran and you flower pl OM water gave 2PL Nastaran and you watered the plants.

PAGE 293

293 Table A 5. Continued item stimuli gloss 9 shoma v nsrin ba hm hmkari krd in you and Nas rin with each other cooperating did 2PL Nasrin and you were cooperating with each other. 10 mlihe v mn shishe ha ro pak krd im Maliheh and I glass pl OM cleaning did 1PL Maliheh and I cleaned the glasses. 11 shoma v xahr etan ashpazi krd in you and sister your cooking did 2PL you and your sister cooked well. 12 s yrus v mn qsm xord im Syrus and I swearing collided 1PL Cyrus and I swore. 13 shoma v neda nqashi ha ro rng krd in you and Neda drawing pl OM painting did 2PL you and Neda painted the drawings. 14 mn v zhale tnab bazi krd im I and Zhale rope playing did 1PL Zhaleh and I skipped the rope. 15 shoma v mrziye shaxe ha ye ezafi ro qeichi krd in you and Marziye branch pl EZ extra OM scissors did 2PL Marziyeh and you cut the extra branches. 16 shenid m ke to v amir namzd krd in heard I that you and Amir engaging did 2PL I heard that Amir and you engaged.

PAGE 294

294 Table A 5. Continued item stimuli gloss 17 mn v sheida ba hm avaz xund im I and Sheida together song sang 1PL Sheida and I sang together. 18 mehran v shoma mntqe ro nqshe brdari krd in Mehran and you area OM map draw did 2PL Mehran and you draw a map of the area. 19 mn v shhram ketab ro xolase krd im I and Shahram book OM summarizing did 1PL Shahram and I summarized the book. 20 behruz v to brname ha ro rdif krd in Behruz and you program pl OM organizi ng did 2PL Behrooz and you organized the programs. 21 mn v shirin ahu ha ro tmasha krd im I and Shirin deer pl OM watching did 1PL Shirin and I watched the deer 22 hmid goft ke reza v to ketab ha ro jabeja krd in Hamid told that Reza and you book pl OM moving did 2PL Hamid told that Reza and yo u moved the books. 23 shoma v shr tkalif ro njam dad in you and Sahar homework OM doing gave 2PL Sahar and you did the homework 24 mn v bhmn hyat ro jaru krd im I and Bahman yard OM sweeping did 1PL Bahman and I swept the yard.

PAGE 295

295 Table A 5. Continued item stimuli gloss 25 frshad v shoma dustan e smimi e mn bud in Farshad and you friend EZ intimate EZ I were 2PL Farshad and you were my intimate friends. 26 mn v friba sox nrani ro gush krd im I and Fariba lecture OM ear did 1PL Fariba and I listened to the lecture. 27 bizhn v mn dr yek sherkt kar krd im Bijan and I in one company working did 1PL Bijan and I worked in the same company. 28 mn v mhin be hm komk krd im I and Mahin to each other helping did 1PL Mahin and I helped each other. 29 sa'id v mn braye grdesh brnamerizi krd im Saeed and I for picni c planning did 1PL Saeed and I planed for the picnic. 30 roya v to divar ha ro taz'in krd in Roya and you wall pl OM decorating did 2P L Roya and you decorated the walls. 31 to v hmd dr istgah e otobus mo'tl shod in you and Ahmad in stop EZ bus waited became 2pl Ahmad and you waited at the bus stop. 32 mn v madr m hm qide bud im I and mother my same idea were 1PL my mother and I were (people of) the same ideas

PAGE 296

296 Table A 5. Continued item stimuli gloss 33 mn v ladn ba hm tse zd im I and Ladan together sneezing hit 1pl Ladan and I sneezed together. 34 behnam v to be bche ha qza dad in Behnam and you to child pl food gave 2PL Behnam and you fed the children. 35 mn v sho'le dr xabgah e doxtr an kar krd im I and Sholeh in dorm EZ girl pl working did 1PL Sholeh and I worked in the girls dormitory. 36 mehrane goft ke behnam v shoma qbul shod in Mehrane to ld that Behnam and you admitting became 2pl Mehrane told that Behnam and you were admitted. 37 frhad v mn ba hm kuh nvrdi ro tjrobe krd im Far had and I together mountain climbing OM did 1PL Farhad and I experienced the mountain climbing together. 38 shokufe v to dr porozhe kar krd in Shokufe and you in project working did 2PL Shokufeh and you worked in the project. 39 mnuchehr v mn hmdigr ro dust dasht im Manuchehr and I each other OM loving dad 1PL Manoochehr and I loved each other. 40 shrare v shoma dr mosabeqe sherkt krd in Sharareh and you in competition p articipating did 2PL Sharareh and you participated in the competition.

PAGE 297

297 Table A 5. Continued item stimuli gloss 41 mehrnush jayeze gereft mehrnush prize took.3SG Mehrnush won the prize 42 mnizhe z movfqiyt ha ysh trif krd Manizhe from success pl her describing did.3SG Manizhe talked of her successes. 43 prvin se ta z she'r ha ro hefz krd Parvin three of poem pl OM memorizing did.3SG Parvin memorized three of the poems. 44 mina ba bhram drs xund Mina with Bahram lesson read.3SG Mina studied with Bahram. 45 xosro srma xord Khosro cold collided.3SG Bahman got a cold. 46 hsn tarix e iran ro motale'eh krd Hasan h istory EZ Iran OM studying did.3SG Hasan studied the history of Iran. 47 siyavsh braye sherkt konnde gan tar zd Siyash for participant pl tar hit.3SG Siyavash played Tar music for the participants. 48 xosro nstrn ro seda krd Khosro Nastaran OM calling did.3SG Khosro called Nastaran.

PAGE 298

298 Table A 5. Continued item stimuli gloss 49 shahin f'aliyt e xub i ro tjrobe krd Shahin activity EZ good one OM experiencing did.3SG Shahin experienced a good activity. 50 minu braye mosafert amade shod Minu for T rip preparing became 3SG Minu prepared for the trip. 51 morqabi ye bozorg I dr an trf e daryache lane dasht duck EZ big one at that side EZ lake nest had.3SG a big duck had nest at the other side of the lake. 52 prstu mahi polo thiye krd Parastu fish rice preparing did.3SG Parastu cooked rice and fish. 53 hesabdar e sherkt drxast e bazneshstegi krd accountant EZ company applying EZ retirement did.3SG the accountant of the company applied for retirement. 54 prnde be juje ha ysh qza dad bird to chick pl her food gave.3SG the bird fed her chicks. 55 kurosh se bxsh z neveshte ha ro trjome krd Kurosh thre e chapter of text pl OM translating did.3SG Kurosh translated three chapters of the texts. 56 pedr e krm mriz bud father EZ Akram sick was.3SG Akram's father w as sick

PAGE 299

299 Table A 5. Continued item stimuli gloss 57 mhshid negran e nomre ha ysh bud Mahshid worry EZ grade pl her was.3SG Mahshid worried for her grades. 58 sediqe kaqz ha ro rng krd Sediqe paper pl OM painting did.3SG Sedighe painted the papers. 59 shbnm dr mosabeqe sherkt krd Shabnam in competition participating did.3SG Shabnam partici pated in the competition. 60 li joqrafiya ro dust dasht Ali geography OM love had.3SG Ali loved geography. 61 hdis ba li movafeq bud Hadis with Ali agree was.3SG Hadis agreed with Ali. 62 omid be esfhan sfr krd Omid to Esfahan traveling did.3SG Omid traveled to Isfahan. 63 mnizhe z hmkelasi ysh komk gereft Manizhe from classmate her help took.3SG Manizheh asked help from her classmate. 64 prstar daru ro be mriz dad nurse medicine OM to patient gave.3SG the nurse gave the medicines to the patient.

PAGE 300

300 Table A 5. Continued item stimuli gloss 65 jmshid do name braye golnush post krd Jamshid two letter for Golnush mailing did.3SG Jamshid mailed two letters to Goli. 66 nrges z nomre ha ysh razi bud Narges from score PL her satisfy was.3SG Narges was satisfied with her scores. 67 fsane braye madr sh yek kif entexab krd Afsane for mother her one bag choosing did.3SG Afasaneh chose a bag for her mother. 68 hedye goft ke fshin xeili tlash krd Hedye told that Afshin a lot struggling did.3SG Hedye told tha t Afshin struggled a lot. 69 fereidun kar e digri peida krd Fereidun job_EZ another finding did.3SG Fereidun found another job. 70 hom ayun goft ke ezafe kari ha ro prdaxt krd n Homayun told that overtime pl OM paying did 3PL Homayoon told that they have paid the overtimes. 71 mehran hdf esh ro tozih dad Mehran goal his OM explaining gave.3SG Mehran explained his goal to his father. 72 mhtab sa't esh ro peida krd Mahtab watch his OM finding did.3SG Mahtab found her watc h.

PAGE 301

301 Table A 5. Continued item stimuli gloss 73 shhram ezdevaj krd Shahram marrying did.3SG Shahram married. 74 mahi e kuchk dr rudxane shena krd fish EZ little in river swimming did.3SG the little fish swam in the river. 75 prvane dr jshnvare jayeze gereft Parvaneh in festival prize took.3SG Pa rvaneh got a prize in the festival. 76 syrus z mostjer sh shekayt krd Cyrus from lessee his complaining did.3SG cyrus complained of his lessee. 77 frzane ketab e roman ro nqd krd Farzaneh book EZ novel OM critisizing did.3SG Farzaneh critisized the novel. 78 susan mlafe ha ro otu krd Susan sheet pl OM ironing did.3SG Susan ironed the bed sheets. 79 prisa ba mehrdad bhs krd Parisa with Mehrdad discussing did.3SG Parisa discussed with Mehrdad. 80 mrym ba mehri telefon i sohbt krd Maryam with Mehri telephone on talking did.3SG Maryam talked with Mehri on phone.

PAGE 302

302 Table A 6. List of the filler stimuli for judgme nt experiment (Condition: present) item stimul i gloss 1 mn v friborz msa'el e riyazi ro hl mi kon im I and Fariborz problem EZ math OM solving IND do 1PL Fariborz and I solve math problem. 2 mn v ehsan hva ye srd ro dust dar im I and Ehsan weather EZ cold OM loving have 1PL Ehsan and I like cold weather. 3 sho ma v mhshid emtehan dar in you and Mahshid exam have 2PL Mahshid and you have an exam. 4 shoma v hsn ab e mahi ha ro vz mi kon in you and Hasan water EZ fish pl OM changing IND do 2PL Hasan and you change the water of the fish bowl. 5 sara v mn z muze didn mi kon im Sara and I from museum visiting IND do 1PL Sara and I visit the museum. 6 shhab v shoma drs ha ro morur mi kon in Shahab and you lesson pl OM r eviewing IND do 2PL Shahab and you review the lessons. 7 kaveh v mn hmkelas hst im Kaveh and I classmate be 1PL Kaveh and I are classmates. 8 nstrn v shoma g ol ha ro ab mi d in Nastaran and you flower pl OM water IND give 2PL Nastaran and you water the plants.

PAGE 303

303 Table A 6. Continued item stimuli gloss 9 shoma v nsrin ba hm hmkari mi kon in you and Nasrin with each other cooperating IND do 2PL Nasrin and you cooperate with each other. 10 mlihe v mn shishe ha ro pak mi kon im Maliheh and I glass pl OM cleaning IND do 1PL Maliheh and I clean the glasses. 11 shoma v xahr etan ashpazi mi kon in you and sister your cookin g IND do 2PL you and your sister cook. 12 syrus v mn qsm mi xor im Syrus and I swearing IND collide 1PL Syrus and I are going to swear. 13 shoma v neda nqashi ha ro rng mi kon in you and Neda drawing pl OM painting IND do 2PL you and Neda paint the drawings. 14 mn v zhale tnab bazi mi kon im I and Zhale rope playing IND do 1PL Zhaleh and I skip the rope. 15 shoma v mrziye shaxe ha ye ezafi ro qeichi mi kon in you and Marziye branch pl EZ extr a OM scisors IND do 2PL Marziyeh and you cut the extra branches. 16 shenid m ke to v amir namzd mi kon in heared I that you and Amir engaging IND do 2PL I heard that Amir and you engage.

PAGE 304

304 Table A 6. Continued item S timuli gloss 17 mn v sheida ba hm avaz mi xun im I and Sheida with together song IND sing 1PL Sheida and I sing together. 18 mehran v shoma mntqe ro nqshe brdari mi kon in Mehran and you area OM map drawing IND do 2PL Mehran and you draw a map of the area. 19 mn v shhram se fsl e ketab ro xolase mi kon im I and Shahram three chapter book OM summarizing IND do 1PL Shahram and I are going to summarize three chapters of the book. 20 behruz v to brname ha ro rdif mi kon in Behruz and you program pl OM organizing IND do 2PL Behr ooz and you organize the programs. 21 mn v shirin ahu ha ro tmasha mi kon im I and Shirin deer pl OM watching IND do 1PL shirin and I watch the deer. 22 hmid goft ke reza v to ketab ha ro jabeja mi kon in Hamid told that Reza and you book pl OM moving IND do 2PL Hamid told that Reza and you move the books. 23 shoma v shr tkalif ro njam mi deh in you and Sahar homework OM doing IND give 2PL Sahar and you do the homework. 24 mn v b hmn hyat ro jaru mi kon im I and Bahman yard OM sweeping IND do 1PL Bahman and I sweep the yard.

PAGE 305

305 Table A 6. Continued item S timuli gloss 25 frshad v shoma dustan e smimi e mn hst in Farshad and you friend EZ intimate EZ I be 2PL Farshad and you are my intimate friends. 26 mn v friba soxnrani ro gush mi kon im I and Fariba lecture OM ear IND do 1PL Fariba and I listen to the lecture. 27 bizhn v mn dr yek sherkt kar mi kon im Bijan I I in one company working IND do 1PL Bijan and I work in the same company. 28 mn v mhin be hm komk mi kon im I and Mahin to each other helping IND do 1PL Mahin and I help each other. 29 sa'id v mn braye grdesh brnamerizi mi kon im Saeed and I for picnic plani ng IND do 1PL Saeed and I plan for the picnic. 30 roya v to divar ha ro taz'in mi kon in Roya and you wall pl OM decorating IND do 2PL Roya and you decorate the walls. 31 to v hmd dr istgah e otobus mo'tl mi sh in you and Ahmad in stop EZ bus waited IND become 3pl Ahmad and you wait at the bus stop. 32 mn v madr m hm qide hst im I and mother my same idea be 1PL my mother and I are (people) of the same ideas.

PAGE 306

306 Table A 6. Continued item stimuli glo ss 33 mn v ladn ba hm tse mi zn im I and Ladan together sneezing IND -hit 1PL Ladan and I sneeze at the same time. 34 behnam v to be bche ha qza mi d in Behnam and you to child pl food IND give 2PL Behnam and you feed the children. 35 mn v sho'le dr xabgah e doxtr an kar mi kon im I and Sholeh in dorm EZ girl pl working IND do 1PL Sholeh and I work in the girls dormitory. 36 mehrane goft ke behnam v shoma qb ul mi sh in Mehrane told that Behnam and you admitting IND become 3PL Mehrane told that Behnam and you will be admitted. 37 frhad v mn ba hm kuh nvrdi ro tjrobe mi kon im Farhad and I together mountain climbing OM experiencing IND do 1PL Farhad and I experience the mountain climbing together. 38 shokufe v to dr porozhe kar mi kon in Shokufe and you in project working IND do 2PL Shokufeh and you work in the project. 39 mnuchehr v mn hmdigr ro dust dar im Manuchehr and I each other OM loving have 1PL Manoochehr and I love each other. 40 shrare v shoma dr mosabeqe sherkt mi kon in Sharareh and you in competition participating IND do 2PL Sharareh and you participate in the competition.

PAGE 307

307 Table A 6. Continued item stimuli gloss 41 mehrnush J ayeze mi gir e mehrnush P rize IND take 3SG Mehrnush wins the prize. 42 mnizhe z movfqiyt ha ysh trif mi kon e Manizhe from success pl her explaining IND do 3SG Manizhe talks of her successes. 43 prvi se ta z she'r ha ro hefz mi kon e Parvin three from poem pl OM memorizing IND do 3SG Parvin memorizes three of the poems. 44 mina ba bhram drs mi xun e Mina with Bahram lesson IND sing 3SG Mina studies with Bahram. 45 xosro S rma mi xor e Khosro C old IND collide 3SG Bahman gets a cold. 46 hsn tarix e iran ro motale'eh mi kon e Hasan history EZ Iran OM studying IND do 3SG Hasan studies the history of Iran. 47 siyavsh B raye sherkt konnde gan tar mi zn e Siyash for part icipant pl Tar IND hit 3SG Siyavash plays Tar music for the participants. 48 xosro nstrn ro seda mi kon e Khosro Nastaran OM calling IND do 3SG Khosro calls Nastaran.

PAGE 308

308 Table A 6. Continued item stimuli gloss 49 shahin f'aliyt e xub i ro tjrobe mi kon e Shahin activity EZ good one OM experiencing IND do 3SG Shahin experiences a good activity. 50 minu braye mosafert amade mi sh e Minu for T rip preparing IND become 3SG Minu prepares for the trip. 51 morqabi ye bozorg I dr an trf e darya che lane dar e duck EZ big one at that side EZ lake nest have 3SG A big duck has nest at the other side of the lake. 52 prstu mahi polo thiye mi kon e Parastu fish ric e preparing IND do 3SG Parastu cooks rice and fish. 53 hesabdar e sherkt drxast e bazneshstegi mi kon e accountant EZ company applying EZ retirement IND do 3SG the accountant of the c ompany applies for retirement. 54 prnde be juje ha ysh qza mi de e bird to chick pl her food IND give 3SG the bird feeds her chicks. 55 kurosh se b xsh z neveshte ha ro trjome mi kon e Kurosh three chapter of text pl OM translating IND do 3SG Kurosh translates three chapters of the texts. 56 pedr e krm mriz e father EZ Akram sick be.3SG Akram's father is sick.

PAGE 309

309 Table A 6. Continued item stimuli gloss 57 mhshid negran e nomre ha sh e Mahshid worried EZ grade pl her be.3SG Mahshid worries for her grades 58 sediqe kaqz ha ro rng mi kon e Sediqe paper pl OM painting IND do 3SG Sedighe paints the papers. 59 shbnm dr mosabeqe sherkt mi kon e S habnam in competition participating IND do 3SG Shabnam participates in the competition. 60 li joqrafiya ro dust dar e Ali geography OM loving have 3SG Ali loves geography. 61 hdis ba li movafeq e Hadis with Ali agreeing be.3SG Hadis agrees with Ali. 62 omid be esfhan sfr mi kon e Omid to Esfahan travelling IND do 3SG Omid travels to Isfahan. 63 mnizhe z hmkelasi ysh komk mi gir e Manizhe from classmate her help IND take 3SG Manizheh asks help from her classmate. 64 prstar daru ro be mriz mi de e nurse medicine OM to patient IND give 3SG the nurse gives the medicines to the patient.

PAGE 310

310 Table A 6. Continued item stimuli gloss 65 jmshid do name braye golnush post mi kon e Jamshid two letter for Golnush mailing IND do 3SG Jamshid mails two letters to Goli. 66 nrges z nomre ha ysh razi ye Narges from grade PL her satisfied be.3SG Narges is satisfied with her grades. 67 fsane braye madr sh yek kif entexab mi kon e Afsane for mother her one bag choosing IND do 3SG Afsaneh chooses a bag for her mother. 68 hedye goft ke fshin xeili tlash mi kon e Hedye told that Afshin a lot effort IND do 3SG He dye told that Afshin is working hard. 69 fereidun kar e digri peida mi kon e Fereidun job EZ another finding IND do 3SG Fereidun finds another job. 70 homayun goft ke ezafe kari ha ro prdaxt mi kon n Homayun told that overtime pl OM paying IND do 3PL Homayoon told that they are going to pay overtimes. 71 mehran hdf esh ro tozih mi de e Mehran goal his OM explaining IND give 3SG Mehran explains his goal. 72 mhtab sa't esh ro peida mi kon e Mahtab watch his OM finding IND do 3SG Mahtab finds her watch.

PAGE 311

311 Table A 6. Continued item stimuli gloss 73 shhram E zdevaj mi kon e Shahram M arrying IND do 3SG Shahram is going to marry. 74 mahi e kuchk dr rudxane shena mi kon e fish EZ little in river swimming IND do 3SG the little fish is swimming in the river. 75 prvane dr jshnvare jayeze mi gir e Parv aneh in festival prize IND take 3SG Parvaneh is going to receive a prize in the festival. 76 syrus z mostjer sh shekayt mi kon e Cyrus from lessee his complaining IND do 3SG Cyrus complains of his lessee. 77 frzane ketab e roman ro nqd mi kon e Farzaneh book EZ novel OM critisizing IND do 3SG Farzaneh reviews the novel. 78 susan mlafe ha ro otu mi kon e Susan sheet pl OM ironing IND do 3SG Susan irons the bed sheets. 79 prisa ba mehrdad bhs mi kon e Parisa with Mehr dad discussing IND do 3SG Parisa discusses with Mehrdad. 80 mrym ba ba mehri telefon i sohbt mi kon e Maryam with with Mehri telephone on talking IND do 3SG Maryam talks with Mehri on phone.

PAGE 312

312 APPENDIX B GLOSS VERB TYPE VERB TENSE EXPERIME N T

PAGE 313

313 Table B 1. List of the target stimuli for Verb Type Verb Tense experiment No. VT preamble gloss possible target response 1 A tnab ha mi xor e mi xor n xord xord n rope pl IND collide 3SG IND collide 3PL collide.Past.3SG collide.Past 3PL the ropes knot ... the ropes became knotted. 2 A znjir ha mi xor e mi xor n xo rd xord n chain pl IND collide 3SG IND collide 3PL collide.Past.3SG collide.Past 3PL the chains twist ... the chains became twisted. 3 A dr ha mi xor e mi xor n xord xord n door pl IND collide 3SG IND collide 3PL collide.Past.3SG collide.Past 3PL the doors became painted. 4 A frsh ha mi xor e mi xor n xord xord n rug pl IND collide 3SG IND collide 3PL collide.Past.3SG collide.Past 3PL the rugs dust ... the rugs became dusty. 5 A ketab xune ha mi gir e mi gir n gereft gereft n library pl IND take 3SG IND take 3PL take.Past.3SG take.Past 3PL the libraries flourishing ... the libraries flourished. 6 A dftr ha mi mun e mi mun n mund mund n notebook pl IND remain 3SG IND remain 3PL remain .Past.3SG remain.Past 3PL the notebooks remaining ... the notebooks remained.

PAGE 314

314 Table B 1. Continued No. VT preamble gloss possible target response 7 A sqf ha mi riz e mi riz n rixt rixt n roof pl IND collape 3SG IND collape 3PL collapse.Past.3SG collapse.Past 3PL the roofs collapsed. 8 A gerdu ha mi xor e mi xor n xord xord n walnut pl IND collide 3S G IND collide 3PL collide.Past.3SG collide.Past 3PL the walnuts spining ... the walnuts spun. 9 A bste ha mi xor e mi xor n xord xord n package pl IND collide 3SG IND collid e 3PL collide.Past.3SG collide.Past 3PL the packages returning ... the packages returned. 10 A kfsh ha mi xor e mi xor n xord xord n shoe pl IND collide 3SG IND collide 3PL collide.Pa st.3SG collide.Past 3PL the shoes sliding ... the shoes slided. 11 A mive ha mi r e mi r n rft rft n fruit pl IND go 3SG IND go 3PL go.Past.3SG go.Past 3PL the fruits wasting the fruits were wasted. 12 A brg ha mi xor e mi xor n xord xord n leaf pl IND collide 3SG IND collide 3PL collide.Past.3SG collide.Past 3PL the leaves swaying ... the leaves swayed.

PAGE 315

315 Table B 1. Continued No. VT preamble gloss possible target response 13 A sng ha mi r e mi r n rft rft n stone pl IND go 3SG IND go 3PL go.Past.3SG go.Past 3PL the stones sin king ... the stones sunk. 14 A taxte ha mi mun e mi mun n mund mund n board pl IND remain 3SG IND remain 3PL remain.Past.3SG remain.Past 3PL the boards floating ... the boards were floating. 15 B lubia ha mi zn e mi zn n zd zd n bean pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the beans sprouting ... the beans sp routed. 16 B mixk ha mi kon e mi kon n krd krd n carnation IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the carnations flowering ... the carnations flowered. 17 B saqe ha mi d e mi d n dad dad n stem pl IND give 3SG IND give 3PL give.Past.3SG give.Past 3PL the stems produced leaf ... the stems budded. 18 B dune ha mi kon e mi kon n krd krd n seed pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the seeds growing ... the seeds grew.

PAGE 316

316 Table B 1. Continued No. VT preamble gloss possible target response 19 B tufan h a mi shin e mi shin n neshst neshst n storm pl IND sit 3SG IND sit 3PL sit.Past.3SG sit.Past 3PL the storms subsiding ... the storms subsided. 20 B xmir ha mi a d mi a n am d amd n dough pl IND come 3SG IND come 3PL come.Past.3SG come.Past 3PL the dough raising .... the dough raised. 21 B dst ha mi kon e mi kon n krd krd n hand pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the hands swelling ... the hands swelled. 22 B zxm ha mi xor e mi xor n xord xord n sore pl scabbing over IND collide 3SG IND collide 3PL collid e.Past.3SG collide.Past 3PL the sores scabbing over ... the sores scabbed over. 23 B pa ha mi zn e mi zn n zd zd n foot pl IND strike 3SG IND strike 3PL strike.Past.3SG strike. Past 3PL the feet blistering ... the feet blistered.

PAGE 317

317 Table B 1. Continued No. VT preamble gloss possible target response 24 B ahn ha mi zn e mi zn n zd zd n iron pl IND strik e 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the irons rusting ... the irons rusted. 25 B shaxe ha mi kon e mi kon n krd krd n branch pl IND DO 3SG IND DO 3PL do.Past .3SG do.Past 3PL the branches blossoming ... the branches bloss o med. 26 B jvane ha mi kon e mi kon n krd krd n twig pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the twi gs budding ... the twigs budded. 27 B xiyar ha mi zn e mi zn n zd zd n cucumber pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the cucumbers molding ... the cucumbers molded. 28 B gndom ha mi kon e mi kon n krd krd n wheat pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the wheat taking root ... the wheat took root. 2 9 C lamp ha mi d e mi d n dad dad n lamp pl IND give 3SG IND give 3PL give.Past.3SG give.Past 3PL the lamps lighting ... the lamps gave light.

PAGE 318

318 Table B 1. Continued No. VT preamble glos s possible target response 30 C ruzname ha mi kon e mi kon n krd krd n newspaper pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the newspapers rustling ... the newspapers rust led. 31 C ash ha mi zn e mi zn n zd zd n porridge pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the porridge bubbling ... the porridge bubbled. 32 C atshfeshan ha mi kon e mi kon n krd krd n volcano pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the volcanoes gushing ... the volcanoes gushed. 33 C boshke ha n mi kon e mi kon n krd krd n barrel pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the barrels oozing ... the barrels oozed. 34 C seke ha mi kon e mi kon n krd krd n coin pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the coins jingling ... the coins jingled. 35 C bad ha mi kesh e mi kesh n keshid keshid n wind pl IND pull 3 SG IND pull 3PL pull.Past.3SG pull.Past 3PL the winds moaning ... the winds moaned.

PAGE 319

319 Table B 1. Continued No. VT preamble gloss possible target response 36 C cheraq ha mi zn e mi zn n zd zd n light pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the lights flickering ... the lights flickered. 37 C mte ha mi kon e mi kon n krd krd n drill pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the drills buzzing ... the drills buzzed. 38 C livan ha mi zn e mi zn n zd zd n glass pl IND strike 3SG IND strike 3PL str ike.Past.3SG strike.Past 3PL the glasses gleaming ... the glasses gleamed. 39 C sim ha mi zn e mi zn n zd zd n cord pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3 PL the electrical cords sparking ... the electrical cords sparked. 40 C rre ha mi kon e mi kon n krd krd n saw pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the sa ws making noise ... the saws made noise. 41 C pnjre ha mi kon e mi kon n krd krd n window pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the windows rattling ... the windows rattled.

PAGE 320

320 Table B 1. Continued No. VT preamble gloss possible target response 42 C durbin ha mi zn e mi zn n zd zd n camera pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the cameras flashing ... the cameras flashed. 43 D ferestnde ha mi kon e mi kon n krd krd n beacon pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the beacons transmitting ... the beacons transmitted 44 D durbin ha mi gir e mi gir n gereft gereft n camera pl IND take 3SG IND take 3PL take.Past.3SG take.Past 3PL the camer as photographing ... the cameras photographed. 45 D helikupter ha mi zn e mi zn n zd zd n helicopter pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the hel icopters rotating ... the helicopters rotated. 46 D qayeq ha mi kon e mi kon n krd krd n boat pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the boats escaping ... the boats escaped. 47 D radiyo ha mi d e mi d n dad dad n radio pl IND give 3SG IND give 3PL give.Past.3SG give.Past 3PL the radios informing ... the radios informed.

PAGE 321

321 Table B 1. Continued No. VT preamble gloss possible target response 48 D asansor ha mi r e mi r n rft rft n elevator pl IND go 3SG IND go 3PL go.Past.3SG go.Past 3PL the elevators going up ... the elevators went up. 49 D taksi ha mi zn e mi zn n zd zd n taxi pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the taxies turning around ... the ta xies turned around. 50 D mte ha mi kon e mi kon n krd krd n drill pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the drills made a hole. 51 D vanet bar ha mi kon e mi kon n krd krd n truck pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the trucks smuggling ... the trucks smuggled. 52 D tank ha mi kon e mi kon n krd krd n tank pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the tanks attacking ... the tanks attacked. 53 D pnke ha mi kon e mi kon n krd krd n fan pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the fans cooling ... the fans cooled.

PAGE 322

322 Table B 1. Continued No. VT preamble gloss possible target response 54 D hva peima ha prvaz mi kon e mi kon n krd krd n airplane pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the airplanes flying ... the airplanes flew. 55 D bolndgoo ha mi kon e mi kon n krd k rd n loudspeaker pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the loudspeakers announcing ... the loudspeakers announced. 56 D radar ha mi kon e mi kon n krd krd n radar pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the radars identifying ... the radars identified. 57 E baqlava ha mi zn e mi zn n zd zd n baklava pl winki IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the baklavas winking ... the baklavas winked. 58 E qablame ha mi kon e mi kon n krd krd n pot pl IND DO 3 SG IND DO 3PL do.Past.3SG do.Past 3PL the pots cheating ... the pots cheated. 59 E medad ha mi kon e mi kon n krd krd n pencil pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the pencils frowning ... the pencils frowned.

PAGE 323

323 Table B 1. Continued No. VT preamble gloss possible target response 60 E jaru ha mi r e mi r n rft rft n broom pl IND go 3SG IND go 3PL go.Pa st.3SG go.Past 3PL the brooms way ... the brooms walked. 61 E divar ha mi zn e mi zn n zd zd n wall pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the walls whistling ... the walls whistled. 62 E rusri ha mi zn e mi zn n zd zd n scarf pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the scarves smiling .. the scarves smiled. 63 E tr ha mi kon e mi kon n krd krd n perfume pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the perfumes crying ... the perfumes cried. 64 E lule ha mi kon e mi kon n krd krd n pipe pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the pipes exercising ... the pipes exercised. 65 E x odkar ha mi zn e mi zn n zd zd n pen pl IND strike 3SG IND strike 3PL strike.Past.3SG strike.Past 3PL the pens kneeling ... the pens kneeled.

PAGE 324

324 Table B 1. Continued No. VT preamble gloss possibl e target response 66 E bolndgu ha mi kon e mi kon n krd krd n loudspeaker pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the loudspeakers agreeing ... the loudspeakers agreed. 67 E goje ha mi kon e mi kon n krd krd n tomato pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the thinking ... the thought. 68 E noxod ha mi kon e mi kon n krd krd n pea pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the peas quarreling ... the peas quarreled 69 E kamputer ha mi kon e mi kon n krd krd n computer pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the computers bowing ... the computers bowed. 70 E pnjre ha mi kon e mi kon n krd krd n window pl IND DO 3SG IND DO 3PL do.Past.3SG do.Past 3PL the windows exercising ... the windows exercised. 71 F golabi ha mi sh e mi sh n shod shod n pear pl IND become 3SG IND become 3PL beco me.Past.3SG become.Past 3PL the pears were eaten.

PAGE 325

325 Table B 1. Continued No. VT preamble gloss possible target response 72 F jurab ha mi sh e mi sh n shod shod n so ck pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the socks ... worn the socks were worn. 73 F dastan ha mi sh e mi sh n shod shod n story pl IND become 3 SG IND become 3PL become.Past.3SG become.Past 3PL the stories ... read the stories were read. 74 F dstmal ha mi sh e mi sh n shod shod n handkerchief pl IND become 3SG IND beco me 3PL become.Past.3SG become.Past 3PL the handkerchiefs ... bought the handkerchiefs were bought. 75 F damn ha mi sh e mi sh n shod shod n skirt pl IND become 3SG IND become 3PL be come.Past.3SG become.Past 3PL the skirts ... sewn the skirts were sewn. 76 F tpe ha mi sh e mi sh n shod shod n hill pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the hills ... seen the hills were seen. 77 F boshqab ha mi sh e mi sh n shod shod n plate pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the plates ... washe d the plates were washed.

PAGE 326

326 Table B 1. Continued No. VT preamble gloss possible target response 78 F qndan ha mi sh e mi sh n shod shod n sugar bowl pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the sugar bowls ... broken the sugar bowls were broken. 79 F chmedan ha mi sh e mi sh n shod shod n suitcase pl IND become 3SG IND become 3PL be come.Past.3SG become.Past 3PL the suitcases ... sold the suitcases were sold. 80 F tut frngi ha mi sh e mi sh n shod shod n strawberry pl IND become 3SG IND become 3PL become .Past.3SG become.Past 3PL the strawberries ... picked the strawberries were picked. 81 F zhaket ha mi sh e mi sh n shod shod n sweater pl IND become 3SG IND become 3PL become.Past .3SG become.Past 3PL the sweaters ... knitted the sweaters were knitted. 82 F eink ha mi sh e mi sh n shod shod n eye glass pl IND become 3SG IND become 3PL become.Past.3SG be come.Past 3PL the eye glasses ... counted the eye glasses were counted. 83 F ngoshtr ha mi sh e mi sh n shod shod n ring pl IND become 3SG IND become 3PL become.Past.3SG bec ome.Past 3PL the rings ... sent the rings were sent.

PAGE 327

327 Table B 1. Continued No. VT preamble gloss possible target response 84 F hvij ha mi sh e mi sh n shod shod n carrot pl IND bec ome 3SG IND become 3PL become.Past.3SG become.Past 3PL the carrots ... cooked the carrots were cooked. 85 G miz ha ye n bud bud n table pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the tables ... wooden the tables were wooden. 86 G medad ha e n bud bud n pencil pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the pencils ... red the pencils were red. 87 G sib ha e n bud bud n apple pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the apples ... crisp the apples were crisp. 88 G goje ha s n bud bud n tomato pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the ... fresh the were fresh. 89 G xane ha e n bud bud n house pl be PR ES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the houses ... far the houses were far.

PAGE 328

328 Table B 1. Continued No. VT preamble gloss possible target response 90 G nrde ha ye n bud bud n fence p l be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the fences ... metallic the fences were metallic. 91 G pele ha ye n bud bud n stair pl made of stone be PRES 3SG be PRES 3PL be. Past.3SG be.Past 3PL the stairs ... made of stone the stairs were made of stone. 92 G sndli ha ye n bud bud n chair pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL th e chairs ... Leather the chairs were leather. 93 G qir ha e n bud bud n tar pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the tar ... black the tar was black. 94 G mdrese ha e n bud bud n school pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the schools ... near the schools were near. 95 G txt ha e n b ud bud n bed pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the beds ... big the beds were big.

PAGE 329

3 29 Table B 1. Continued No. VT preamble gloss possible target response 96 G brf ha e n bud bud n snow pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the snow ... White the snow were white. 97 G gushvare ha e n bud bud n ear ring pl be PRES 3 SG be PRES 3PL be.Past.3SG be.Past 3PL the ear rings ... beautiful the ear rings were beautiful. 98 G shal ha ye n bud bud n shawl pl be PRES 3SG be PRES 3PL be.Past.3SG be.Past 3PL the shawls ... silk the shawls were silk. 99 H zmin ha mi sh e mi sh n shod shod n land pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL turated the lands were saturated. 100 H sa't ha mi sh e mi sh n shod shod n watch pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the watches... lost the wa tches were lost. 101 H lke ha mi sh e mi sh n shod shod n stain pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the stains ... wiped the stains were wiped.

PAGE 330

330 Tab le B 1. Continued No. VT preamble gloss possible target response 102 H sbzi ha mi sh e mi sh n shod shod n herb pl withered/faded IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the herbs .. withered/faded the herbs were withered/faded. 103 H qayeq ha mi sh e mi sh n shod shod n boat pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the boats ... sunk the boats were sunk. 104 H badkonak ha mi sh e mi sh n shod shod n balloon pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the balloons ... floated away the balloons were floated away. 105 H balon ha mi sh e mi sh n shod shod n hot air balloon pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the hot air balloons ... floated aw ay the hot air balloons are/were floated away. 106 H nbar ha mi sh e mi sh n shod shod n warehouse pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the warehouses .. ruined the warehouses were ruined. 107 H seke ha mi sh e mi sh n shod shod n coin pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the coins ... found the coins were found.

PAGE 331

331 Table B 1. Continued No. VT preamble gloss possible target response 108 H siman ha mi sh e mi sh n shod shod n cement pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the cement ... set the sement became set. 109 H bstni ha mi sh e mi sh n shod shod n ice cream pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the ice cream ... melted the ice cream was melted. 110 H rng ha mi sh e mi sh n shod shod n paint pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the paint ... dry the paint dried. 111 H varzeshgah ha mi sh e mi sh n shod shod n stadium pl IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the stadiums ... ruined the stadiums were ruined 112 H toxm e morqha mi sh e mi sh n shod shod n egg EZ hen IND become 3SG IND become 3PL become.Past.3SG become.Past 3PL the eggs ... end the eggs ran out.

PAGE 332

332 Table B 2. List of the filler items for Verb Type Verb Tense experiment No. preamble gloss Possible response 1 mn v friborz msa'el e riyazi krd im mi kon im I and Fariborz problem s EZ math OM did 1pl IND do 1pl Fariborz and I solved the math problems. 2 mn v ehsan hva ye srd dasht im dar i m I and Ehsan weather EZ cold OM had 1pl have 1pl Ehsan and I loved cold weather. 3 shoma v mhshid dad in mi deh in you and Mahshid gave 2pl IND give 2pl Mahshid and you took the exam. 4 shoma v hsn ab e mahi ha krd in mi kon in you and Hasan water EZ fish pl OM did.2pl IND do.2pl Hasan and you changed the water of the fish bowl. 5 sara v mn z muze krd im mi kon im Sara and I from museum did 1pl IND do 1pl Sara and I visiting ..... the museum Sara and I visited the museum. 6 shhab v sho ma drs ha krd in mi kon in Shahab and you lesson pl OM did.2pl IND do.2pl Shahab and you reviewed the lessons.

PAGE 333

333 Table B 2. Continued No. preamble gloss Possible response 7 kaveh v Mn bud im hst im Kaveh and I were 1pl be 1pl Kaveh and I were classmates. 8 nstrn v shoma gol ha ro a dad in mi deh in Nastaran and you flower pl OM gave 2pl IND give 2pl Nastaran and you watered the plants. 9 shoma v nsrin ba hm krd in mi kon in you and Nasrin with each other did.2pl IND do.2pl Nasrin and you cooperated with each other. 10 mlihe v mn shishe ha ro krd im mi kon im Maliheh and I glass pl OM did.1pl IND do.1pl Maliheh and I cleaned the glasses. 11 shoma v xahr etan krd in mi kon in you and sister your did.2pl IND do.2pl you and your sister cooked. 12 syrus v mn xord im mi xor im Syrus and I collided.1pl IND collide.1pl Syrus and I swore

PAGE 334

334 Table B 2. Continued No. preamble gloss possible respond 13 shoma v neda nqashi ha krd in mi kon in you and Neda drawing pl OM did.2pl IND do.2pl you and Neda painted the drawings. 14 mn v zhale tnab krd im mi kon im I and Zhale rope did.1pl IND do.1pl Zhaleh and I skipping Zhaleh and I skipped the rope. 15 sh oma v mrziye shaxe ha ye ezafi krd in mi kon in you and Marziye branch pl EZ extra OM scisors ... did.2pl IND do.2pl Marziyeh and you cutting the extra branches. 16 shenid m ke to v ami r krd in mi kon in heard I that you and Ami r did.2pl IND do.2pl I heard that Amir and you engage d. 17 mn v sheida ba hm xund im mi xun im I and Sheida together sang 1pl IND sing 1pl Sheida and I singing..... together Sheida and I song together 18 mehran v shoma mntqe ro nqshe krd in mi kon in Mehran and you area OM map did.2pl IND do.2pl Mehran and you drew a map of the area.

PAGE 335

335 Table B 2. Continued No. preamble gloss possible respond 19 mn v shhram ketab ro krd im mi kon im I and Sh ahram book OM did.1pl IND do.1pl Shahram and I summarizing .....the book Shahram and I summarized the book. 20 behruz v to brname ha ro krd in mi kon in Behruz and you program pl OM did.2pl IND do.2pl Behrooz and you organized the programs. 21 mn v shirin ahu ha ro krd im mi kon im I and Shirin deer pl OM did.1pl IND do.1pl Shirin and I watching..... the deer Shirin and I watched the deer. 22 hmid goft ke reza v t o ketab ha krd in mi kon in Hamid told that Reza and you book pl OM did.2pl IND do.2pl Hamid told that Reza and you moved the books. 23 shoma v shr tkalif dad in mi deh in you and Sahar homework OM gave.2pl IND give.2pl homework Sahar and you did the ho mework 24 mn v bhmn hyat krd im mi kon im I and Bahman yard OM did.1pl IND do.1pl Bahman and I swept the yard

PAGE 336

336 Table B 2. Continued No. preamble gloss possible respond 25 frshad v shoma dustan e smi mi e bud in hst in Farshad and you friend EZ intimate EZ were 2pl be 2pl F Farshad and you were my intimate friends. 26 mn v friba soxnrani krd im mi kon im I and Fariba lecture OM did.1pl IND do.1pl Fariba and I listened to the lecture. 27 bizhn v mn dr yek sherkt krd im mi kon im Bijan and I in one company did.1pl IND do.1pl Bijan and I worked in the same company. 28 mn v mhin be hm krd im mi kon im I and Mahin to each other did.1pl IND do.1pl Mahin and I helping ..... each other Mahin and I helped each other. 29 sa'id v mn braye grdesh brnamerizi ... krd im mi kon im Saeed and I for picnic did.1pl IND do.1pl Saeed and I planning..... for the picnic Saeed and I planed for the picnic. 30 roya v to divar ha ro krd in mi kon in Roya and you wall pl OM decorating ... did.2pl IND do.2pl Roya and you decorated the walls.

PAGE 337

337 Table B 2. Co ntinued No. preamble gloss possible respond 31 to v hmd dr istgah e shod in mi sh in you and Ahmad in stop EZ became.2pl IND become.2pl Ahmad and you waited at the bus stop. 32 mn v madr m bud im hst im I and mother my were 1pl be 1pl my mother and I had the same ideas. 33 mn v ladn ba hm krd im mi kon im I and Ladan together did.1pl IND do.1pl Ladan and I sneezed at the same time. 34 behnam v to be bche ha dad in mi deh in Behnam and you to child pl gave 2pl IND give 2pl children Behnam and you fed the children. 35 mn v sho'le dr xabgah e doxtr krd im mi kon im I and Sholeh in dorm EZ girl pl working .. did.1pl IND do.1pl Sholeh and I worked in the girls dormitory. 36 mehrane goft ke behnam v shod in mi sh in Mehrane told th at Behnam and you admitting became.2pl IND become.2pl Mehrane told that Behnam and you admitting Mehrane told that Behnam and you admitted

PAGE 338

338 Table B 2. Continued No. P reamble gloss possible respond 37 frhad v mn kuh nvrdi ro tjrobe ... krd im mi kon im Farhad and I mountain climbing OM did.1pl IND do.1pl Farhad and I experienced mountain climbing together. 38 shokufe v to dr porozhe krd in mi kon in Shokufe and you in project did.2pl IND do.2pl Shokufeh a Shokufeh and you worked on the project. 39 mnuchehr v mn hmdigr ro dust ... dasht im dar im Manuchehr and I each other OM lovi ng ... had 1pl have 1pl Manoochehr and I loved each other. 40 shrare v shoma dr mosabeqe krd in mi kon in Sharareh and you in competition did.2pl IND do.2pl Sharareh and you participated in the competition. 41 mehrnush gereft mi gir e mehrnush got.3SG IND get 3SG Mehrnush got the prize. 42 mnizhe z movfqiyt ha ysh krd mi kon e Manizhe from success pl her did.3SG IND do 3SG Mani Manizhe talked of her successes.

PAGE 339

339 Table B 2. Continued No. P reamble gloss possible respond 43 prvin se ta z she'r ha ro krd mi kon e Parv in three from poem pl OM did.3SG IND do 3SG Parvin memorized three of the poems. 44 mina ba xund mi xun e Mina with studied.3SG IND study 3SG Mina studied with Bahram. 45 xosro xord mi xor e Khosro collided.3SG IND collid.3SG Bahman got a cold. 46 hsn tarix e iran ro krd mi kon e Hasan history EZ Iran OM did.3SG IND do 3SG Hasan study Hasan studied the history of Iran. 47 siyavsh braye sherkt konnde gan zd mi zn e Siyash for participant pl stroke.3SG IND strike 3SG Siyavas Siyavash played Tar music for the participants. 48 xosro nstrn ro krd mi kon e Khosro Nastaran OM did.3SG IND do 3SG Khos Khosro called Nastaran.

PAGE 340

340 Table B 2. Continued No. preamble gloss possible respond 49 shahin f'aliyt e xub i ro krd mi kon e Shahin activity EZ good one OM did.3SG IND do 3SG Shahin experienced a good activity. 50 minu braye mosafert shod mi sh e Minu for trip became.3SG IND become 3SG Minu prepared for the trip. 51 morqabi ye bozorg I dr an trf e dryache lane ... dasht dar e duc k EZ big one at that side EZ lake nest ... had.3SG have 3SG a big duck had nest at the other side of the lake. 52 prstu mahi polo krd mi kon e Parastu fish did.3SG IND do 3SG Parastu prepared rice and fish. 53 hesabdar e sherkt drxast e krd mi kon e accountant EZ company applying EZ did.3SG IND do 3SG the accountant of the company applied for retirement. 54 prnde be juje ha ysh dad mi deh bird to chick pl her gave.3SG IND give.3SG the bird fed her chicks

PAGE 341

341 Table B 2. Continued No. preamb le gloss possible respond 55 kurosh se bxsh z neveshte ha krd mi kon e Kurosh three chapter of text pl OM did.3SG IND do 3SG three chapters of the texts Kurosh translated three chapters of the texts. 56 pedr e bud e father EZ was.3SG be.3SG Akram' s father was sick. 57 mhshid negran e nomre ha bud e Mahshid worrying EZ grade pl was.3SG be.3SG Mahshid worried for her grades. 58 sediqe kaqz ha ro krd mi kon e Sediqe paper pl OM did.3SG IND do 3SG Sedighe painted the papers. 59 shbnm dr mosabeq e krd mi kon e Shabnam in competition did.3SG IND do 3SG Shabnam participated in the competition. 60 li joqrafiya dasht dar e Ali geography OM had.3SG have 3SG Ali loved geography.

PAGE 342

342 Table B 2. Continued No. preamble gloss possible respond 61 hdis ba bud e Hadis with was.3SG be.3SG Hadis agreed with Ali. 62 Omid be krd mi kon e Omid to did.3SG IND do 3SG Omid traveled to Isfahan. 63 mnizhe z hmkelasi ysh gereft mi gir e Manizhe from classmat e her took.3SG IND take.3SG Manizheh asked for help from her classmate. 64 prstar daru ro be dad mi deh nurse medicin e OM to gave.3SG IND give.3SG the nurse gave the medicines to the patient. 65 jmshid do name braye golnush krd mi kon e Jamshid two letter for Golnush did.3SG IND do 3SG Jamshid mailed two letters to Goli. 66 nrges z nomre ha ysh bud e Nar ges from score pl her was.3SG be.3SG Narges was satisfied with her scores.

PAGE 343

343 Table B 2. Continued No. preamble gloss possible respond 67 fsane braye madr sh yek krd mi kon e Afsane for mother her one bag selecting ... did.3SG IND do 3SG Afasaneh selected a bag for her mot her. 68 hedye goft ke fshin xeili krd mi kon e Hedye told that Afshin a lot did.3SG IND do 3SG Hedye told that Afs hin tried a lot. 69 fereidun kar e digri krd mi kon e Fereidun job_EZ another did.3SG IND do 3SG Fereidun found another job. 70 homayun goft ke ezafe kari ha ro krd n mi kon n Homayun told that overtime pl OM did.3pl IND do.3pl Homayoon told that (they) pay Homayoon told that (they) paid the overtimes. 71 mehran hdf esh ro dad mi deh Mehran goal his OM gave.3SG IND give.3SG Me Mehran explained his goal. 72 mhtab sa't esh ro krd mi kon e Mahtab watch his OM did.3SG IND do 3SG Mahtab found her watch.

PAGE 344

344 Table B 2. Continued No. preamble gloss possible respond 73 shhram krd mi kon e Shahram did.3SG IND do 3SG 74 mahi e kuchk dr rudxane krd mi kon e fish EZ little in river swimming did.3SG IND do 3SG the little fish swimming the little fish swam in t he river. 75 prvane dr jshnvare gereft mi gir e Parvaneh in festival took.3SG IND take 3SG Parvaneh took a prize in the festival. 76 syrus z mostjer sh krd mi kon e Cyrus from lessee his did.3SG IND do 3SG Cyrus complained of his lessee. 77 frzane ketab e roman krd mi kon e Farzaneh book EZ novel OM did.3SG IND do 3SG Farzaneh reviewed the novel. 78 susan mlafe ha ro krd mi kon e Susan sheet pl OM did.3SG IND do 3SG Susan ironed the bed sheets.

PAGE 345

345 Table B 2. Continued No. pre amble gloss possible respond 79 prisa ba mehrdad krd mi kon e Parisa with Mehrdad did.3SG IND do 3SG Parisa discussed with Mehrdad. 80 mrym ba mehri telefon i krd mi kon e Maryam with Mehri telephone on did.3SG IND do 3SG Maryam talked with Mehri on phone Maryam talk ed with Mehri on phone.

PAGE 346

346 APPENDIX C GLOSS UNITY EXPERIMENT

PAGE 347

347 Table C 1. List of the stimuli for Unity experiment No. str Unity NO P reamble gloss possible target response 1a Inch IND PL dr ha ye daxel e char chub h a rngi shod /shod n door PL EZ inside EZ framework PL colorful became.3SG /became 3PL the doors in the frameworks were mottled. 1b Inch Unity PL dr ha ye nzdik e char chub ha rngi shod /shod n door PL EZ near EZ framework PL colorful became.3SG /became 3PL the doors near the frameworks were mot tled. 1c Inch Unity SG dr ha ye nzdik e char chub rngi shod /shod n door PL EZ near EZ framework colorful became.3SG /became 3PL the doors near the framework mottled the d oors near the framework were mottled. 2a Inch IND PL chador ha ye ruy e mashin ha pare shod /shod n cover PL EZ on EZ car PL torn became.3SG /became 3PL the covers on the cars were torn. 2b Inch Unity PL chador ha ye nzdik e mashin ha pare shod /shod n cover PL EZ near EZ car PL torn became.3SG /became 3PL the covers near the covers near the cars were torn. 2c Inch Unity SG chador ha ye nzdik e mashin pare shod /shod n cover PL EZ near EZ car torn became.3SG /became 3PL the the covers near the car were torn.

PAGE 348

348 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 3a Inch IND PL howle ha ye ruy e qolab ha xis shod /shod n towel PL EZ on EZ hook PL wet became.3SG /became 3PL the towels on the hooks became wet. 3b Inch Unity PL howle ha ye nzdik e qolab ha xis shod /shod n towel PL EZ near EZ hook PL wet became.3SG /became 3PL the towels near the hooks became wet. 3c Inch Unity SG howle ha ye nzdik e qola b xis shod /shod n towel PL EZ near EZ hook PL wet became.3SG /became 3PL the towels near the hook became wet. 4a Inch IND PL dstmal ha y e ruy e boshqab ha ksif shod /shod n tissue PL EZ on EZ plate PL dirty became.3SG /became 3PL T he tissues on the plates became dirty. 4b Inch Unity PL dstmal ha ye nzdik e boshqab ha ksif shod /shod n tissue PL EZ near EZ plate PL dirty became.3SG /became 3PL T he tissues near the plates became dirt y. 4c Inch Unity SG dstmal ha ye nzdik e boshqab ksif shod /shod n tissue PL EZ near EZ plate dirty became.3SG /became 3PL T he tissu es near the plate became dirty.

PAGE 349

349 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 5a Inch IND PL komod ha ye daxel e otaq ha qdimi shod /shod n dresser PL EZ inside EZ room PL old became.3SG /became 3PL the dressers inside the rooms became old. 5b Inch Unity PL komod ha ye birun e otaq ha qdimi shod /shod n dresser PL EZ outside EZ room PL old became.3SG /became 3PL the dressers outside the rooms became old. 5c Inch Unity SG komod ha ye birun e otaq qdimi sh od /shod n dresser PL EZ outside EZ room old became.3SG /became 3PL the dressers outside the room became old. 6a Inch IND PL jurab ha ye daxel e kfsh ha kohne shod /shod n sock PL EZ inside EZ shoes PL old became.3SG /became 3PL T he socks in the shoes became old. 6b Inch Unity PL jurab ha ye nzdik e kfsh ha kohne shod /shod n sock PL EZ near EZ shoes PL old became.3SG /became 3PL T he socks near the shoes became old. 6c Inch Unity SG jurab ha ye nzdik e kfsh kohne shod /shod n sock PL EZ near EZ shoes old became.3SG /became 3PL T he socks near the shoe became old.

PAGE 350

350 Table C 1. Conti nued No. str uc Unity NO preamble gloss possible target response 7a Inch IND PL mobayl ha ye daxel e kif ha xamush shod /shod n cell phone PL EZ inside EZ bag PL off became.3SG /became 3PL T he cell phones in the bags wer e turned off. 7b Inch Unity PL mobayl ha ye nzdik e kif ha xamush shod /shod n cell phone PL EZ near EZ bag PL off became.3SG /became 3PL T he cell phones near the bags were turned off. 7c Inch Unity SG mobayl ha ye nzdik e kif xamush shod /shod n cell phone PL EZ near EZ bag off became.3SG /became 3PL T he cell phones near the bag were turned off. 8a Inch IND PL mlafe ha ye ruy e txt ha lke dar shod /shod n sheet PL EZ on EZ bed PL s tained became.3SG /became 3PL the sheets on the beds were stained. 8b Inch Unity PL mlafe ha ye kenar e txt ha lke dar shod /shod n sheet PL EZ beside EZ bed PL 350 tained became.3SG /became 3PL T he sheets beside the beds were stained 8c Inch Unity SG mlafe ha ye kenar e txt lke dar shod /shod n sheet PL EZ beside EZ bed 350 tained became.3SG /became 3PL T he sheets beside the bed were stained.

PAGE 351

351 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 9a Inch IND PL lamp ha ye ruy e miz ha rowshn shod /shod n lamp PL EZ on EZ desk PL on became.3SG /became 3PL the T he lamps on the desks were turned on. 9b Inch Unity PL lamp ha ye nzdik e miz ha rowshn shod /shod n lamp PL EZ near EZ desk PL on became.3SG /became 3PL T he lamps near the desks were turned on. 9c Inch Unity SG lamp ha ye nzdik e miz rowshn shod /shod n lamp PL EZ near EZ desk on became.3SG /beca me 3PL T he lamps near the desk were turned on. 10a Inch IND PL mlaqe ha ye daxel e qablme ha daq shod /shod n ladle PL EZ inside EZ pot PL hot became.3SG /became 3PL T he ladles in the pots became hot. 10b Inch Unity PL mlaqe ha ye nzdik e qablme ha daq shod /shod n ladle PL EZ near EZ pot PL hot became.3SG /became 3PL T he ladles near the pots became hot. 10c Inch Unity SG mlaqe ha ye nzdik e qablme daq shod /shod n ladle PL EZ near EZ pot hot became.3SG /became 3PL T he ladles near the pot became hot.

PAGE 352

352 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 11a Inch IND PL ayene ha ye daxel e qab ha keder shod /shod n mirror PL EZ inside EZ frame PL opaque became.3SG /became 3PL the mirrors in the frames became opaq ue. 11b Inch Unity PL ayene ha ye nzdik e qab ha keder shod /shod n mirror PL EZ near EZ frame PL opaque became.3SG /became 3PL O paque. T he m irrors near the frames became opaque. 11c Inch Unity SG ayene ha ye nzdik e qab keder shod /shod n mirror PL EZ near EZ frame opaque became.3SG /became 3PL O paque. T he mirrors near the frame became opaque. 12a Inch IND PL estxr ha ye daxel e baq ha por shod /shod n pool PL EZ inside EZ garden PL full became.3SG /became 3PL the T he pools in the gardens became full. 12b Inch Unity PL estxr ha ye nzdik e baq ha por shod /shod n pool PL EZ near EZ garden PL full became.3SG /became 3PL T he pools near the gardens became full. 12c Inch Unity SG estxr ha ye nzdik e baq por shod /shod n pool PL EZ near EZ garden full beca me.3SG /became 3PL T he pools near the garden became full.

PAGE 353

353 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 13a Pass IND PL tel eviziun ha ye ruy e miz ha dozdid e shod /shod n television PL EZ on EZ table PL stole pp became.3SG /became 3PL T he TVs on the tables were stolen. 13b Pass Unity PL televiziun ha ye nzdik e miz ha dozdid e shod /shod n television PL EZ near EZ table PL stole pp became.3SG /became 3PL T he TVs near the tab les were stolen. 13c Pass Unity SG televiziun ha ye nzdik e miz dozdid e shod /shod n television PL EZ near EZ table stole pp became.3SG /became 3PL T he TVs near the table were stolen. 14a Pass IND PL quri ha ye ruy e smavr ha foruxt e shod /shod n tea pot PL EZ on EZ samovar PL sold pp became.3SG /became 3PL the t T he teapots on the samovars were sold. 14b Pass Unity PL quri ha ye nzdik e smavr ha foruxt e shod /shod n tea pot PL EZ near EZ samovar PL sold pp became.3SG /became 3PL T he teapots near the samovars were sold. 14c Pass Unity SG quri ha ye nzdik e smavr foruxt e shod /shod n tea pot PL EZ near EZ samovar sold pp became.3SG /became 3PL T he teapots near the samovar were sold.

PAGE 354

354 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 15a Pass IND PL ngoshtr ha ye daxel e jbe ha ferestad e shod /shod n ring PL EZ inside EZ box PL sent pp became.3SG /became 3PL T he rings inside the box es were sent. 15b Pass Unity PL ngoshtr ha ye nzdik e jbe ha ferestad e shod /shod n ring PL EZ near EZ box PL sent pp became.3SG /became 3PL T he rings near the boxes were sent. 15c Pass Unity SG ngoshtr ha ye nzdik e jbe ferestad e shod /shod n ring PL EZ near EZ box sent pp became.3SG /became 3PL the rings near T he rings near the box were sent. 16a Pass IND PL eink ha ye tu ye jeld ha shomord e shod /shod n glasses PL EZ in EZ case PL counted pp became.3SG /became 3PL T he glasses in the cases were counted. 16b Pass Unity PL eink ha ye nzdik e jeld ha shomord e shod /shod n glasses PL EZ near EZ case PL count ed pp became.3SG /became 3PL T he glasses near the cases were counted. 16c Pass Unity SG eink ha ye nzdik e jeld shomord e shod /shod n glasses PL EZ near EZ case counted pp became.3SG /became 3PL T he glasses near the case were counted.

PAGE 355

355 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 17a Pass IND PL xane ha ye daxel e mzr e ha saxt e shod /shod n house PL EZ inside EZ farm PL built pp became.3SG /became 3PL T he houses in the farms were built. 17b Pass Unity PL xane ha ye nzdik e mzr e ha saxt e shod /shod n house PL EZ near EZ farm PL built pp became.3SG /became 3PL T he houses near the farms were built. 17c Pass Unity SG xane ha ye nzdik e mzr e saxt e shod /shod n house PL EZ near EZ farm built pp became.3SG /became 3PL the houses near the f T he houses near the farm were built. 18a Pass IND PL dstgire ha ye ruy e dr ha shekst e shod /shod n handle PL EZ on EZ door PL broke pp became.3SG /became 3PL T he handles on the doors were broken. 18b Pass Unity PL dstgire ha ye nzdik e dr ha shekst e shod /shod n handle PL EZ near EZ door PL brok e pp became.3SG /became 3PL T he handles near the doors were broken. 18c Pass Unity SG dstgire ha ye nzdik e dr shekst e shod /shod n handle PL EZ near EZ door broke pp became.3SG /became 3PL T he handles near the door were broken.

PAGE 356

356 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 19a Pass IND PL kise ha ye daxel e stl ha rixt e shod /shod n bag PL EZ inside EZ bucket PL poured pp became.3SG /became 3PL T he bags in the buckets were pou red. 19b Pass Unity PL kise ha ye nzdik e stl ha rixt e shod /shod n bag PL EZ near EZ bucket PL poured pp became.3SG /became 3PL T he ba gs near the buckets were poured. 19c Pass Unity SG kise ha ye nzdik e stl rixt e shod /shod n bag PL EZ near EZ bucket poured pp became.3SG /became 3PL T he bags near the bucket were poured. 20a Pass IND PL frsh ha ye daxel e rahrow ha shost e shod /shod n rug PL EZ inside EZ corridor PL washed pp became.3SG /became 3PL t T he rugs inside the corridors were washed. 20b Pass Unity PL frsh ha ye birun e rahrow ha shost e shod /shod n rug PL EZ outside EZ corridor PL w ashed pp became.3SG /became 3PL T he rugs outside the corridors were washed. 20c Pass Unity SG frsh ha ye birun e rahrow shost e shod /shod n rug PL EZ outside EZ corridor washed pp became.3SG /became 3PL T he rugs outside the corridor were washed.

PAGE 357

357 Table C 1. Continued No. str uc Unity NO preamble gloss possible target res ponse 21a Pass IND PL fenjan ha ye ruy e nlbki ha brdasht e shod /shod n cup PL EZ on EZ saucer PL took pp became.3SG /became 3PL T he cups on the saucers wer e taken. 21b Pass Unity PL fenjan ha ye nzdik e nlbki ha brdasht e shod /shod n cup PL EZ near EZ saucer PL took pp became.3SG /became 3PL T he cups near the saucers were taken. 21c Pass Unity SG fenjan ha ye nzdik e nlbki brdasht e shod /shod n cup PL EZ near EZ saucer took pp became.3SG /became 3PL the c T he cups near the saucer were taken. 22a Pass IND PL nx ha ye tu ye suzn ha borid e shod /shod n string PL EZ in EZ needle PL cut pp became.3SG /became 3PL T he strings in the needles were cut. 22b Pass Unity PL nx ha ye nzdik e suzn ha borid e shod /shod n string PL EZ near EZ needle PL cut p p became.3SG /became 3PL T he strings near the needles were cut. 22c Pass Unity SG nx ha ye nzdik e suzn borid e shod /shod n string PL E Z near EZ needle cut pp became.3SG /became 3PL T he strings near the needle were cut.

PAGE 358

358 Table C 1. Continued No. str uc Unity NO preamble gloss possible target response 23a Pass IND PL negin ha ye ruy e ngoshtr ha psndid e shod /shod n gem PL EZ on EZ ring PL chose pp became.3SG /became 3PL T he gems on the rings were chosen. 23b Pass Unity PL negin ha ye nzdik e ngoshtr ha psndid e shod /shod n gem PL EZ near EZ ring PL chose pp became.3SG /became 3PL T he gems near the rings were chosen. 23c Pass Unity SG negin ha ye nzdik e ngoshtr psndid e shod /shod n gem PL EZ near EZ ring chose pp became.3SG /became 3PL the gems near the rin T he gems near the ring were chosen. 24a Pass IND PL keravat ha ye ruy e pirahn ha xrid e shod /shod n tie PL EZ on EZ shirt PL bought pp became.3SG /became 3PL T he ties on the shirts were bought. 24b Pass Unity PL keravat ha ye nzdik e pirahn ha xrid e shod /shod n tie PL EZ near EZ shirt PL bought pp b ecame.3SG /became 3PL T he ties near the shirts were bought. 24c Pass Unity SG keravat ha ye nzdik e pirahn xrid e shod /shod n tie PL EZ near EZ shirt bought pp became.3SG /became 3PL T he ties near the shirt were bought.

PAGE 359

359 Table C 2. List of the filler items for Unity & Concreteness experiment item preamble gloss possible respo nse 1 susan mlafe ha ro out krd Susan sheet PL OM iron did.3SG Susan ironed the sheets. 2 mn v shole dr xabgahe doxt ran kar krd im I and Shole in dormitory girls working did.1PL Shole and I worked in the girls dormitory. 3 s'id v m n braye grdesh brname rizi krd im Sa'id and I for picnic planning did.1PL Sa'id and I planed for the picnic. 4 mn v bhmn hyat ro jaru krd im I and Bahman yard OM sweeping did.1PL Bahman and I swept the yard. 5 prisa ba mehrdad bhs krd Parisa with Mehrdad arguing did.3SG Parisa argued with Mehrdad. 6 shokufe v to dr porozhe sherkt krd in Shokufe and you in project participating did .2PL Sholkufe and you participated in the project.

PAGE 360

360 Table C 2. Continued item preamble gloss possible response 7 sirus z mostjer sh shekayt krd Sirus from tenant his complaining did.3SG Sirus complained form his tenant. 8 shrare v shoma dr mosabeqe sherkt krd in Sharare and yo u in competition participating did.2PL Sharare and you participated in the competition. 9 bizhn v mn dr yek sherkt kar k rd im Bizhan and I in one company working did.1PL Bizhan and I worked in a company. 10 frhad v mn kuhnvrdi ro tjrobe krd im Farhad and I mountain climbing OM experiencing did.1PL Farhad and I experienced mountain climbing. 11 mlihe v mn shishe ha ro pak krd im Malihe and I glass PL OM cleaning did.1PL Malihe and I cleaned the glasses. 12 behruz v to brname ha ro rdif kr d in Behruz and you plan PL OM organizing did.2PL Behruz and you organized the plans.

PAGE 361

361 Table C 2. Continued item preamble gloss possible response 13 mahi kuchik dr rudxune shena krd fish small in river swimming did.3SG the little fish swam in the river. 14 shoma v xahr etan ashpzi krd in yo u and sister your cooking did.2PL you and your sister cooked. 15 jmshid do name braye golnush post krd Jamshid two letters for Golnush posting did.3SG Jamshid posted two letters for Golnush. 16 shoma v neda nqashi ha ro rng krd in you and Neda drawing PL OM painting did.2PL you and Neda painted the drawings. 17 mrym ba mehri telefon i sohbt krd Maryam with Mehri telephone with talking did.3SG Maryam with Mehri talked with telephone. 18 mn v zhale tnab bazi krd im I and zhale rope playing did.1PL Zhale and I skipping Zhale and I skipped rope.

PAGE 362

362 Table C 2. Continued item preamble gloss possible response 19 mehran goft ke behnam v shoma qabul krd in Mehran told that Behnam and you accepti ng did.2PL Mehran told that Behnam and you accepted. 20 shenid m ke to v mir namzd krd in heared 1SG that you and Amir enga ging did.2PL I heard that you and Amir are engaged. 21 behnam v to be bche ha qza dad in Behnam and you to child PL food gave 2PL Behnam and you fed the children. 22 mn v sheida ba hm avaz xand im I and Sheida together singing sang 1PL S heida and I sand together. 23 prvane dr jshnvare jayeze gereft Parvane in festival prize got.3SG Parvane got a prize in the festival. 24 meh rnush jayeze bord mehrnush prize took.3SG Mehrnush took a prize. 25 mn v ladn ba hm tse zd im I and Ladan together sneezing strike.1PL Ladan and Ladan and I sneezed together.

PAGE 363

363 Table C 2. Continued item preamble gloss possible response 26 Mina ba bhram drs xund n Mina with Bhram lesson read 3PL Mina and Mina and Bahram read the lesson. 27 fsane braye madr sh yek kif entexab krd Afsane for mother her one ba choosing did.3SG Afsane chose a bag for her mother. 28 hsn tarix e iran ro motale'e krd hsn history EZ Iran OM studying did.3SG Hasan studied the histor y of Iran. 29 fereidun kar e digri peida krd Fereidun job EZ another finding did.3SG Fereidun found another job. 30 xosro nstrn ro seda krd Xosro Nastaran OM calling did.3SG Xosro called Nastaran. 31 mnuchehr v mn hmdigr ro dust dasht im Manuchehr and I each other OM loving had 1 PL Manucher and I loved each other. 32 minu braye mosafert amade shod Minu for travel preparing become.3SG Minu was prepar ed for the travel.

PAGE 364

364 Table C 2. Continued item preamble gloss possible response 33 mn v friba soxnrani ro gush krd im I and Fariba lecture OM listening did.1PL Fariba and I listening Fariba and I listened to the lecture. 34 prstu mahi polo thiye krd Parastu fish rice preparing did.3SG Parastu prepared fish and ri ce. 35 roya v to divar ha ro tz'in krd in Roya and you wall PL OM decorate did.2PL Roya and you decorated the walls. 36 sirus v mn qsm xord im Sirus and I oath ate 1PL Sirus and I took an oath. 37 sediqe kaqz ha ro rng krd Sediqe paper PL OM paint ing did.3SG sediqe painted the papers. 38 shoma v shr tkalif ro njam dad in you and Sahar assignment OM doing gave 2PL Sah Sahar and you did the assignment.

PAGE 365

365 Table C 2. Continued item preamble gloss possible response 39 shbnm dr mosabeqe sherkt krd Shabnm in competition participating did .3SG Shabnam participated in the competition. 40 mn v mhin be hm komk krd im I and Mahin to each other helping did.1PL Mahin Mahin and I helped each other. 41 prstar daru ro be mriz dad nurse medicine OM to patient gave.3SG The nurse gave the medicine to the patient. 42 hmid goft ke reza va to ketab ha ro jabeja krd in Hamid told that Reza and you book PL OM moving did. 2PL Hamid told tha t Hamid told tha Reza and you changed the place of the books. 43 li joqrafia ro dust dasht Ali geography OM liking had.3SG Ali likin Ali liked Geography. 44 mnizhe z hmkelasi sh komk gereft Manizhe from classmate her help got.3SG Manizhe got help form her classmate.

PAGE 366

366 Table C 2. Continued item preamble gloss possible response 45 mhtab sa't sh ro peida krd Mhtab watch her OM finding did.3SG Mahtab found her watch. 46 frzane ketab e roman ro nqd krd Farzane book EZ novel OM reviewing did.3SG 47 xosro srma xord xosro col d ate.3SG xosro got cold. 48 homayun goft ke ezafe kari ha ro prdaxt krd n Homayun told that overtime PL OM paying did.3PL Homayun told that Homayun told that they paid the overtimes.

PAGE 367

367 APPENDIX D GLOSS CONCRETENESS EXPERIMENT Table D 1 List of the stimuli for Concreteness experiment item structure condition preamble gloss possible target response 1 a inchoative abstract vaqe'iyt ha pdidar shod /shod n reality PL visible became.3SG /became 3PL the realities became visible. 1b inchoative concrete keshti ha pdidar shod /shod n ship PL visible became.3SG /became 3PL the ships became visible. 2a inchoative abstract moskel ha nzdik shod /shod n problem PL close became.3SG /became 3PL the problems became close. 2b inchoative concrete gari ha nzdik shod /shod n cart PL close became.3SG /became 3PL the carts became close. 3a inchoative abstract sxtgiri ha qt' shod /shod n severity PL cut became.3SG /became 3PL the severities were cut. 3b inchoative concrete ab ha qt' shod /shod n water PL cut became.3SG /became 3PL the w ater (in the pipe) was cut.

PAGE 368

368 Table D 1. Continued item structure condition preamble gloss possible target response 4a inchoative abstract mohbt ha km shod /shod n affection PL less became.3SG / became 3PL the affections were decreased. 4b inchoative concrete medad ha km shod /shod n pencil PL less became.3SG /became 3PL ss. the pencils were decreased. 5a inchoative abstract pishrft ha moshxs shod /shod n development PL clear became.3SG /became 3PL the developments became clear. 5b inchoative concrete cheraq ha moshxs shod /shod n light PL clear became.3SG /became 3PL the lights became clear/vi sible. 6a inchoative abstract e'temad ha sost shod /shod n trust PL weak became.3SG /became 3PL the trusts were weaken. 6b inchoative concrete paye ha sost shod /shod n pillar PL weak became.3SG /became 3PL the pillars were weaken. 7a inchoative abstract bavr ha sxt shod /shod n belief PL strong became.3SG /became 3PL the beliefs became strong.

PAGE 369

369 Table D 1. Continued item structure condition preamble gloss possible target response 7b inchoative concrete siman ha sxt shod /shod n cement PL hard/strong became.3SG /became 3PL the cements were set. 8a inchoative abstract doshmni h a napdid shod /shod n animosity PL invisible became.3SG /became 3PL the animosities were vanished. 8b inchoative concrete lamt ha napdid shod /shod n sign PL invisible became.3SG /became 3PL the signs were vanished. 9a inchoative abstract mehrbani ha nmayan shod / shod n kindness PL visible became.3SG /became 3PL the kindnesses became visible. 9b inchoative concrete rud ha nmayan shod /shod n river PL visible became.3SG /became 3PL the rivers became visible. 10a inchoative abstract delsuzi ha ziyad shod /shod n sympathy PL more became.3SG /b ecame 3PL the sympathies increaded. 10b inchoative concrete qayeq ha ziyad shod /shod n boat PL more became.3SG /became 3PL the boats increased.

PAGE 370

370 Table D 1. Continued item structure condition preamble gloss possible target response 11a inchoative abstract budje ha tmam shod /shod n budget PL end became.3SG /became 3PL the budgets ran out. 11b inchoative concrete shokolat ha tmam shod /shod n chocolate PL end became.3SG /became 3PL the choc the chocolates ran out. 12a inchoative abstract drdesr ha peida shod /shod n headache PL visible became.3SG /became 3PL the headaches were appeared. 12b inchoative concrete kelid ha peida shod /shod n key PL visible became.3SG /became 3PL the keys were appeared/found. 13a passive abstract emtiyaz ha foruxt e shod /shod n concession PL sold pp became.3SG /became 3PL the concessions were sold. 13b passive concret e kamyun ha foruxt e shod /shod n truck PL sold pp became.3SG /became 3PL the trucks were sold. 14a passive abstract sxti ha shomo rd e shod /shod n difficulty PL counted pp became.3SG /became 3PL the difficulties were counted.

PAGE 371

371 Table D 1. Continued item structure condition preamble gloss possibl e target response 14b passive concrete xodkar ha shomord e shod /shod n pen PL counted pp became.3SG /became 3PL the pens were counted. 15a passive abstract ziba'i ha did e shod /shod n beauty PL saw pp became.3SG /became 3PL the beauties were seen. 15b passive concrete havapeima ha did e shod /shod n plane PL saw pp became.3SG /became 3PL the planes were seen. 16a passive abstract komk ha ferestad e shod /shod n assistance PL sent pp beca me.3SG /became 3PL the assistances were sent. 16b passive concrete toshk ha ferestad e shod /shod n mattress PL sent pp became.3SG /became 3PL the mattresses were sent. 17a passive abstract hzine ha prdaxt e shod /shod n expenditure PL paid pp became.3SG /became 3PL the expenditur the expenditures were paid. 17b passive concrete surt hesab ha prdaxt e shod /shod n bill PL paid pp became.3SG /became 3PL the bills were paid.

PAGE 372

372 Table D 1. Continued item structure condition P reamble gloss possible target response 18a passive abstract ide ha nevesht e shod /shod n idea PL wrote pp became.3SG /beca me 3PL the ideas were written. 18b passive concrete vrqe ha nevesht e shod /shod n paper PL wrote pp became.3SG /became 3PL en. the papers were written. 19a passive abstract omid ha dad e shod /shod n hope PL gave pp became.3SG /became 3PL the hopes were given. 19 b passive concrete kamputer ha dad e shod /shod n computer PL gave pp became.3SG /became 3PL the computers were given. 20a passive abstract fekr ha pziroft e shod /shod n thought PL accepted pp became.3SG /became 3PL the thoughts were accepted. 20b passive concrete kif ha pziroft e shod /shod n bag PL accepted pp became.3SG /became 3PL the bags were accepted.

PAGE 373

373 Table D 1. Continued item structure condition preamble gloss possible target r esponse 21a passive abstract rftar ha psndid e shod /shod n behavior PL approved pp became.3SG /became 3PL the behaviors were approved. 21b passive concrete lebas ha psndid e shod /shod n clothe PL approved pp became.3SG /became 3PL the clothes were approved/liked. 22a passive abstract fhng ha shenaxt e shod /shod n culture PL knew pp became.3SG /became 3PL the cultures were known. 22b passive concrete srzmin ha shenaxt e shod /shod n land PL knew pp became.3SG /became 3PL the lands were known. 23a passive abstract peiman ha shekst e shod /shod n agreement PL broke pp became.3SG /became 3PL the agreements were broken. 23b passive concrete shishe ha shekst e shod /shod n glass PL broke pp became.3SG / became 3PL the glasses were broken.

PAGE 374

374 Table D 1. Continued item structure condition preamble gloss possible target response 24a passive abstract trh ha xand e shod / shod n plan PL studied pp became.3SG /became 3PL the plans were studied. 24b passive concrete ketab ha xand e shod /shod n book PL studied pp became.3SG /became 3PL the books were studied.

PAGE 375

375 APPENDIX E GLOSS ATTRACTION EXPERIMENT

PAGE 376

376 Table E 1. List of the stimuli for Attraction experiment item animacy NO Preamble gloss possible target response 1a Ani pl an ... taksi ha kudk an ro krd /krd n taxi pl child pl OM did.3SG /did 3PL the taxies moving ... the children' the taxies transported the children. 1b Ani pl h a ... taksi ha kudk ha ro krd /krd n taxi pl child pl OM did.3SG /did 3PL the taxies moving ... the children' the taxies transported the children. 1c A ni S g ... taksi ha kudk ro krd /krd n taxi pl child OM did.3SG /did 3PL the taxies moving ... the child' the taxies transported the child. 1d inani pl ha ... taksi ha jbe ha ro krd /krd n taxi pl box pl OM did.3SG /did 3PL the taxies moving ... the boxes' the taxies transported the boxes. 1e inani sg ... taksi ha Jbe ro krd /krd n taxi pl Box OM did.3SG /did 3PL the taxies moving ... the box' the taxies transported the box.

PAGE 377

377 Table E 1. Continued item a nimacy NO Preamble gloss possible target response 2a ani pl an ... ambulans ha mjruh an ro dad /dad n ambulance pl injured pl OM gave.3SG /gave 3PL injured people the ambulances made a place for the injured people 2b ani pl ha ... ambulans ha mjruh ha ro dad /dad n ambulance pl injured pl OM gave.3SG /gave 3PL th e ambulances the ambulances made a place for the injured people 2c ani sg ... ambulans ha Mjruh ro dad /dad n ambulance pl Injured OM gave.3SG / gave 3PL the ambulances made a place for the injured. 2d inani pl ha ... ambulans ha mlafe ha ro dad /dad n ambulance pl sheet pl OM p gave.3SG /gave 3PL the ambulances made a place for the sheets. 2e inani sg ... ambulans ha Mlafe ro dad /dad n ambulance pl Sheet OM gave.3SG /gave 3PL the ambulances made a place for the sheet.

PAGE 378

378 Table E 1. Continued item animacy NO Preamble gloss possible target response 3a ani pl an ... qolab ha shotor an ro keshid /keshi d n grapnel pl camel pl OM pull.3SG /pull 3PL the grapnels pulled the camels back. 3b ani pl ha ... qolab ha shotor ha ro keshid /keshi d n grapnel pl camel pl OM pull.3SG /pull 3PL the grapnels pulled the camels back. 3c ani sg ... qolab ha Shotor ro keshid /keshi d n grapnel pl Camel OM pull.3SG /pull 3PL the grapnels pulled the camel back. 3d inani pl ha ... qolab ha prde ha ro keshid /keshi d n grapnel pl curtain pl OM pull.3SG /pull 3PL the grapnels pulled the curtains back. 3e inani sg ... qolab ha Prde ro keshid /keshi d n grapnel pl Curtain OM pull.3SG /pull 3PL the grapnels pulled the c urtain back.

PAGE 379

379 Table E 1. Continued item animacy NO Preamble gloss possible target response 4a ani pl an ... fnr ha bazikon an ro krd /krd n spring pl player pl OM did.3SG /did 3PL the springs throwing ... the players' the springs threw the players. 4b ani pl ha ... fnr ha bazikon ha ro krd /krd n spring pl player pl OM did.3 SG /did 3PL the springs throwing ... the players' the springs threw the players. 4c ani sg ... fnr ha Bazikon ro krd /krd n spring pl Player OM did.3SG /did 3PL the springs throwing ... the player' the springs threw the player. 4d inani pl ha ... fnr ha tup ha ro krd /krd n spring pl ball pl OM did.3SG /did 3PL the springs throwing ... the balls' the springs threw the balls. 4e inani sg ... fnr ha tup ro krd /krd n spring pl ball OM did.3SG /did 3 PL the springs throwing ... the ball' the springs threw the ball.

PAGE 380

380 Table E 1. Continued item animacy NO Preamble gloss possible target response 5a ani pl an ... abpash ha rhgozr an ro krd /krd n sprinkler pl passerby pl OM did.3SG /did 3PL the sprinklers wetting ... the passers by' the sprinklers made the passers by wet. 5b ani pl ha ... abpash ha rhgozr ha ro krd /krd n sprinkler pl passerby pl OM did.3SG /did 3PL the sprinklers wetting ... the passers by' the sprinklers made the passers by wet. 5c ani sg ... abpas h ha rhgozr ro krd /krd n sprinkler pl passerby OM did.3SG /did 3PL the sprinklers wetting ... the passer by' the sprinklers made the passer by wet. 5d inani pl ha ... abpash ha dochrxe ha ro krd /krd n sprinkler pl bicycle pl OM did.3SG /did 3PL the sprinklers wetting ... the bicycles' the sprinklers made the bicycles wet. 5e inani sg ... abpash ha dochrxe ro krd /krd n sprinkler pl bicycle OM did.3SG /did 3PL the sprinklers wetting ... the bicycle' the sprinklers made the bicycle wet.

PAGE 381

381 Table E 1 Continued item animacy NO Preamble gloss possible target response 6a ani pl an ... ks ha rusta'i yan ro dad /dad n photo pl villager pl OM gave.3SG /gave 3PL 'the p the photos showed the villagers. 6b ani pl ha ... ks ha rusta'i ha ro dad /dad n photo pl villager pl OM gave.3SG /gave 3PL the photos showed the villagers. 6c ani sg ... ks ha rusta'i ro dad /dad n photo pl villager OM gave.3SG /gave 3PL the photos showed the villager. 6d inani pl ha ... ks ha ayene ha ro dad /dad n photo pl mirror pl OM gave.3SG /gave 3 PL the photos showed the mirrors. 6e inani sg ... ks ha ayene ro dad /dad n photo pl mirror OM gave.3SG /gave 3PL the photos showed the mirrors.

PAGE 382

382 Table E 1. Continued item animacy NO Preamble gloss possible target response 7a ani pl an ... mjle ha dvnde gan ro krd /krd n magazine pl runner pl OM did.3SG /did 3PL the magazines introducing ... the runners' the magazines introduced the runners. 7b ani pl ha ... mjle ha dvnde ha ro krd /krd n magazine pl runner pl OM did.3SG /did 3PL the magazines introducing ... the runners' the magazines introduced the runners. 7c ani sg ... mjle ha dvnde ro krd /krd n magazine pl runner OM did.3SG /did 3PL the magazines introducing ... the runner' the magazines introduced the runner. 7d inani pl ha ... mjle ha gushi ha ro krd /krd n magazine pl cell phone pl OM did.3SG /did 3PL the magazines introducing ... the cell phones' the magazines introduced the c ell phones. 7e inani sg ... mjle ha gushi ro krd /krd n magazine pl cell phone OM did.3SG /did 3PL the magazines introducing ... the cell phone' the ma gazines introduced the cell phone.

PAGE 383

383 Table E 1. Continued item animacy NO Preamble gloss possible target response 8a ani pl an ... list ha brnde gan ro krd /krd n list pl winner pl OM anno did.3SG /did 3PL the lists announcing ... the winners' the lists announced the winners. 8b ani pl ha ... list ha brnde ha ro krd /krd n list pl winner pl OM did.3SG /did 3PL the lists announcing ... the winners' the lists announced the winners. 8c ani sg ... list ha brnde ro krd /krd n list pl winner OM did.3SG /did 3PL the lists announcing ... the winner' the lists announced the winner. 8d inani pl ha ... list ha jayeze ha ro krd /krd n list pl prize pl OM did.3SG /did 3PL the lists announcing ... the prizes' the lists announced the prizes. 8e inani sg ... list ha jayeze ro krd /krd n list pl p rize OM did.3SG /did 3PL the lists announcing ... the prize' the lists announced the prize.

PAGE 384

384 Table E 1. Continued item animacy NO Preamble gloss possible target response 9a ani pl an ... znjir ha gusfnd an ro dasht /dasht n chain pl sheep pl OM kept.3SG /kept 3PL the chains held the sheep. 9b ani pl ha ... znjir ha gusfnd ha ro dasht /dasht n chain pl sheep pl OM kept.3SG /kept 3PL the chains held the sheep. 9c ani sg ... znjir ha gusfnd ro dasht /dasht n chain pl sheep OM kept.3SG /kept 3PL the chains held the sheep. 9d inani pl ha ... znjir ha gari ha ro dasht /dasht n chain pl cart pl OM kept.3SG /kept 3PL the chains held the carts. 9e inani sg ... znjir ha gari r o dasht /dasht n chain pl cart OM kept.3SG /kept 3PL the chains held the cart.

PAGE 385

385 Table E 1. Continued item animacy NO Preamble gloss possible target response 10a ani pl an ... kuler ha mosafer an ro krd /krd n cooler pl passenger pl OM did.3SG /did 3PL the air conditioners cooling ... the passengers' the air conditioners cooled the passenge rs. 10b ani pl ha ... kuler ha mosafer ha ro krd /krd n cooler pl passenger pl OM did.3SG /did 3PL the air conditioners cooling ... the passengers' the air conditi oners cooled the passengers. 10c ani sg ... kuler ha mosafer ro krd /krd n cooler pl passenger OM did.3SG /did 3PL the air conditioners cooling ... the passenger' th e air conditioners cooled the passenger. 10d inani pl ha ... kuler ha otaq ha ro krd /krd n cooler pl room pl OM did.3SG /did 3PL the air conditioners cooling ... the rooms' the air conditioners cooled the rooms. 10e inani sg ... kuler ha otaq ro krd /krd n cooler pl room OM did.3SG /did 3PL the air conditioners cooling ... the room' the air conditioners cooled the room.

PAGE 386

386 Table E 1. Continued item animacy NO Preamble gloss possible target response 11a ani pl an ... vagon ha sb an ro krd /krd n wagon pl horse pl OM carrying did.3SG /did 3PL the wagons carrying ... the horses' the wagons carried the horses. 11b ani pl ha ... vagon ha sb ha ro krd /krd n wagon pl horse pl OM ca rrying did.3SG /did 3PL the wagons carrying ... the horses' the wagons carried the horses. 11c ani sg ... vagon ha sb ro krd /krd n wagon pl horse pl OM carrying did.3SG /did 3PL the wagons carrying ... the horse' the wagons carried the horse. 11d inani pl ha ... vagon ha mojaseme ha ro krd /krd n wagon pl statue pl OM carrying did. 3SG /did 3PL the wagons carrying ... the statues' the wagons carried the statues. 11e inani sg ... vagon ha mojaseme ro krd /krd n wagon pl statue OM carrying did.3SG /did 3PL the wagons carrying ... the statue' the wagons carried the statue.

PAGE 387

387 Table E 1. Continued item animacy NO Preamble gloss possible target response 12a ani pl an ... boxari ha shenagr an ro krd /krd n heater pl swimmer pl OM did.3SG /did 3PL the heaters drying ... the swimmers' the heaters dried the swimmers. 12b ani pl ha ... boxari ha shenag r ha ro krd /krd n heater pl swimmer pl OM did.3SG /did 3PL the heaters drying ... the swimmers' the heaters dried the swimmers. 12c ani sg ... boxari ha shena gr ro krd /krd n heater pl swimmer OM did.3SG /did 3PL the heaters drying ... the swimmer' the heaters dried the swimmer. 12d inani pl ha ... boxari ha qal iche ha ro krd /krd n heater pl rug pl OM did.3SG /did 3PL the heaters drying ... the rugs' the heaters dried the rugs. 12e inani sg ... boxari ha qaliche ro x krd /krd n heater pl rug OM did.3SG /did 3PL the heaters drying ... the rug' the heaters dried the rug.

PAGE 388

388 Table E 1. Continued item animacy NO Preamble gloss possible target response 13a ani pl an ... dstgah ha mriz an ro krd /krd n instrument pl patient pl OM ch did.3SG /did 3PL the instruments checking ... the patients' the instruments checked the patients. 13b ani pl ha ... dstgah ha mriz ha ro krd /krd n instrument pl patient pl OM ch did.3SG /did 3PL the instruments checking ... the patients' the instruments ch ecked the patients. 13c ani sg ... dstgah ha Mriz ro krd /krd n instrument pl Patient OM ch did.3SG /did 3PL the instruments checking ... the patient' the instru ments checked the patient. 13d inani pl ha ... dstgah ha anemone ha ro krd /krd n instrument pl sample p OM ch did.3SG /did 3PL the instruments checking ... the samples' the instruments checked the samples. 13e inani sg ... dstgah ha Nemune ro krd /krd n instrument pl Sample OM ch did.3SG /did 3PL the instruments checking ... the sa mple' the instruments checked the sample.

PAGE 389

389 Table E 1. Continued item animacy NO Preamble gloss possible target response 14a ani pl an ... tofng ha ahu wan ro gereft /gereft n gun pl d eer pl OM took.3SG /took 3PL the guns aimed the deer. 14b ani pl ha ... tofng ha ahu ha ro gereft /gereft n gun pl deer pl OM aiming took.3SG /took 3PL the guns aimed the deer. 14c ani sg ... tofng ha Ahu ro gereft /gereft n gun pl deer OM took.3SG /took 3PL the guns aimed the deer. 14d inani pl ha ... tofng ha balon ha ro gereft /gereft n gun pl hot air balloon pl OM took.3SG /took 3 PL the guns aimed the hot air balloons. 14e inani sg ... tofng ha balon ro gereft /gereft n gun pl hot air balloon OM took.3SG /took 3PL the guns aimed the hot air balloon.

PAGE 390

390 Table E 1. Continued item animacy NO Preamble gloss possible target response 15a ani pl an ... radar ha dos hmn an ro krd /krd n radar pl enemy pl OM did.3SG /did 3PL the radars identifying ... the enemies' the radars identified the enemies. 15b ani pl ha ... radar ha doshmn ha ro krd /krd n radar pl enemy pl OM did.3SG /did 3PL the radars identifying ... the enemies' the radars identified the enemies. 15c ani sg ... radar ha doshmn ro krd /krd n radar pl enemy OM did.3SG /did 3PL the radars identifying ... the enemy' the radars identified the enemy. 15d inani pl ha ... radar ha hvapeima ha ro krd /krd n radar pl plane pl OM did.3SG /did 3PL the radars identifying ... the planes' the radars identified t he planes. 15e inani sg ... radar ha hvapeima ro krd /krd n radar pl plane OM did.3SG /did 3PL the radars identifying ... the plane' the radar s identified the plane.

PAGE 391

391 Table E 1. Continued item animacy NO Preamble gloss possible target response 16a ani pl an ... xtkeshi ha danesh amuz an ro krd /krd n line pl student pl OM sepa did.3SG /did 3PL the lines separating ... the students' the lines separated the students. 16b ani pl ha ... xtkeshi ha danesh amuz ha ro krd /krd n line pl student pl OM did.3SG /did 3PL the lines separating ... the students' the lines separated the students. 16c ani sg ... xtkeshi ha danesh amuz ro krd /krd n line pl student OM did.3SG /did 3PL the lines separating ... the student' the lines separated the student. 16d inani pl ha ... xtkeshi ha xiyaban ha ro krd /krd n line pl street pl OM did.3SG /did 3PL the lines separating ... the streets' the lines separated the streets. 16e inani sg ... xtkeshi ha xiyaban ro krd /krd n line pl street OM did.3SG /did 3PL the lines separating ... the street' the lines separated the street.

PAGE 392

392 Table E 1. Continued item animacy NO Preamble gloss possible target respons e 17a ani pl an ... qltk ha fzanvrd an ro dad /dad n roller pl astronaut pl OM gave.3SG /gave 3PL the rollers shook the astronauts. 17b ani pl ha ... qltk ha fzanvrd ha ro dad /dad n roller pl astronaut pl OM gave.3SG /gave 3PL the rollers shook the astronauts. 17c ani sg ... qltk ha fzanvrd ro dad /dad n roller pl astronaut OM gave.3SG /gave 3PL t he rollers shook the astronaut. 17d inani pl ha ... qltk ha porteqal ha ro dad /dad n roller pl orange pl OM gave.3SG /gave 3PL the rollers shook the oranges. 17e inani sg ... qltk ha porteqal ro dad /dad n roller pl orange OM gave.3SG /gave 3PL the rollers shook the orange.

PAGE 393

393 Table E 1. Continued item animacy NO Preamble gloss possible target response 18a ani pl an ... qayeq ha nhng an ro krd /krd n boat pl whale pl OM did.3SG /did 3PL the boats chasing ... the whales' the boats chased the whales. 18b ani pl ha ... qayeq ha nhng ha ro krd /krd n boat pl whale pl OM did.3SG /did 3PL the boats chasing ... the whales' the boats chased the whales. 18c ani sg ... qayeq ha nhng ro krd /krd n boat pl whale pl OM d id.3SG /did 3PL the boats chasing ... the whale' the boats chased the whale. 18d inani pl ha ... qayeq ha keshti ha ro krd /krd n boat pl ship pl OM did.3 SG /did 3PL the boats chasing ... the ships' the boats chased the ships. 18e inani sg ... qayeq ha keshti ro krd /krd n boat pl ship pl OM did.3SG /did 3PL the boats chasing ... the ship' the boats chased the ship.

PAGE 394

394 Table E 1. Continued item animacy NO Preamble gloss possible target response 19a ani pl an ... brchsb ha mosafer an ro krd /krd n lable pl passenger pl OM did.3SG /did 3PL the labels identifying ... the passengers' the labels identified the passengers. 19b ani pl ha ... brchsb ha mosafer ha ro krd /krd n lable pl passenger pl OM did.3SG /did 3PL the labels identifying ... the passengers' the labels identified the passengers. 19c ani sg ... brchsb ha mosafer ro krd /krd n lable pl passenger OM did.3SG /did 3PL the labels identifying ... the passenger' the labels identified the passenger. 19d i nani pl ha ... brchsb ha karton ha ro krd /krd n lable pl box pl OM did.3SG /did 3PL the labels identifying ... the boxes' the labels identified t he boxes. 19e inani sg ... brchsb ha karton ro krd /krd n lable pl box OM did.3SG /did 3PL the labels identifying ... the box' the labels identi fied the box.

PAGE 395

395 Table E 1. Continued item animacy NO Preamble gloss possible target response 20a ani pl an ... nimkt ha pir mrd an ro krd /krd n seat pl old man pl OM did.3SG /did 3PL the seats staining ... the old men' the seats made a stain on the old men. 20b ani pl ha ... nimkt ha pir mrd ha ro krd /krd n seat pl old man pl OM colo did.3SG /did 3PL the seats staining ... the old men' the seats made a stain on the old men. 20c ani sg ... nimkt ha pir mrd ro krd /krd n seat pl old man O M did.3SG /did 3PL the seats staining ... the old man' the seats made a stain on the old man. 20d inani pl ha ... nimkt ha lebas ha ro krd /krd n seat pl dress pl OM did.3SG /did 3PL the seats staining ... the dresses' the seats made a stain on the dresses. 20e inani sg ... nimkt ha lebas ro krd /krd n s eat pl dress OM did.3SG /did 3PL the seats staining ... the dress' the seats made a stain on the dress.

PAGE 396

396 Table E 1. Continued item animacy NO Preamble gloss possible target response 21a ani pl an ... daru ha bimar an ro dad /dad n medicine pl patient pl OM gave.3SG /gave 3PL the medicines cured the patients. 21b ani pl ha ... daru ha bimar ha ro dad /dad n medicine pl patient pl OM gave.3SG /gave 3PL the medicines cured the patients. 21 c ani sg ... daru ha bimar ro dad /dad n medicine pl patient OM gave.3SG /gave 3PL the medicines cured the patient. 21 d inani pl ha ... daru ha jeraht ha ro dad /dad n medicine pl injury pl OM gave.3SG /gave 3PL the medicines cured th e injuries. 21e inani sg ... daru ha jeraht ro dad /dad n medicine pl injury OM gave.3SG /gave 3PL the medicines cured the injury.

PAGE 397

397 Table E 1. Continued item animacy NO Preamble gloss possible target response 22a ani pl an ... mhlul ha janevr an ro krd /krd n liquid pl animal pl OM did.3SG /did 3PL the liquids polluting ... the animals' the liquids polluted the animals. 22b ani pl ha ... mhlul ha janevr ha ro krd /krd n liquid pl animal pl OM pollute did.3SG /did 3PL the liquids polluting ... the animals' the liquids polluted the animals. 22c ani sg ... mhlul ha janevr ro krd /krd n liquid pl animal OM pol did.3SG /did 3PL the liquids polluting ... the animal' the liquids polluted the animal. 22d inani pl ha ... mhlul ha dryache ha ro krd /krd n liquid p l lake pl OM did.3SG /did 3PL the liquids polluting ... the lakes' the liquids polluted the lakes. 22e inani sg ... mhlul ha dryache ro krd /krd n liq uid pl lake OM did.3SG /did 3PL the liquids polluting ... the lake' the liquids polluted the lake.

PAGE 398

398 Table E 1. Continued item animacy NO Preamble gloss possible target response 23a ani pl an ... pomad ha zxmi yan ro krd /krd n ointment pl wounded pl OM did.3SG /did 3PL the ointments healing ... the wounded people' the ointments healed the wounded people. 23b ani pl ha ... pomad ha zxmi ha ro krd /krd n ointment pl wounded pl OM did.3SG /did 3PL the ointments healing ... the wounded people' the ointments healed the wou nded people. 23c ani sg ... pomad ha zxmi ro krd /krd n ointment pl wounded OM did.3SG /did 3PL the ointments healing ... the wounded person' the ointments hea led the wounded person. 23d inani pl ha ... pomad ha zxm ha ro krd /krd n ointment pl wound pl OM did.3SG /did 3PL the ointments healing ... the wounds' the ointments healed the wounds. 23e inani sg ... pomad ha zxm ro krd /krd n ointment pl wound OM did.3SG /did 3PL the ointments healing ... the wound' the oin tments healed the wound.

PAGE 399

399 Table E 1. Continued item animacy NO Preamble gloss possible target response 24a ani pl an ... tasme ha kuhnvrd an ro dasht /dasht n belt pl mountain climer pl OM h kept.3SG /kept 3PL the belts held the mountain climbers. 24b ani pl ha ... tasme ha kuhnvrd ha ro dasht /dasht n bel t pl mountain climer pl OM kept.3SG /kept 3PL the belts held the mountain climbers. 24c ani S g ... tasme ha kuhnvrd ro dasht /d asht n belt pl mountain climer OM kept.3SG /kept 3PL the belts held the mountain climbers. 24d inani pl ha ... tasme ha asansor ha ro neg dasht /dasht n belt pl elevator pl OM kept.3SG /kept 3PL the belts held the elevators. 24e inani S g ... tasme ha asansor ro neg dasht /dasht n belt pl elevator OM kept.3SG /kept 3PL the belts held the elevator.

PAGE 400

400 Table E 1. Continued item animacy NO Preamble gloss possible tar get response 25a ani pl an ... kabin ha moshtri an ro bord /bord n cabin pl customer pl OM took.3SG /took 3PL the cabins took the customers forward. 25b ani pl ha ... kabin ha moshtri ha ro bord /bord n cabin pl customer pl OM took.3SG /took 3PL the cabins took the cust omers forward. 25c ani sg ... kabin ha moshtri ro bord /bord n cabin pl customer OM took.3SG /took 3PL the cabins took th e customer forward. 25d inani pl ha ... kabin ha vzne ha ro bord /bord n cabin pl weight pl OM took.3SG /took 3PL the cabins took the weights forward. 25e inani sg ... kabin ha vzne ro bord /bord n cabin pl weight OM took.3SG /took 3PL the cabins took the weight forward.

PAGE 401

401 Table E 1. Continued item animacy NO Preamble gloss possible target response 26a ani pl an ... hrom ha gonjeshk an ro krd /krd n lever pl sparrow pl OM did.3SG /did 3PL the levers releasing ... the sparrows' the levers released the sparrows. 26b ani pl ha ... hrom ha gonjeshk ha ro krd /krd n lever pl sparrow pl OM did.3SG /did 3PL the levers releasing ... the sparrows' the levers released the sparrows. 26c ani sg ... hrom ha gonjeshk ro krd /krd n lever pl sparrow OM did.3SG /did 3PL the levers releasing ... the sparrow' the levers released the sparrow. 26d inani pl ha ... hrom ha kise ha ro krd /krd n lever pl sack pl OM did.3SG /did 3PL the levers releasing ... the sacks' the levers released the sacks. 26e inani sg ... hrom ha kise ro krd /krd n lever pl sack OM did.3SG /did 3PL the levers releasing ... the sack' the levers released the sack.

PAGE 402

402 Table E 1. Continued item animacy NO Preamble gloss possible target response 27a ani pl an ... minibus ha aber an ro krd /krd n mini bus pl passerby pl OM did.3SG /did 3PL the mini buses crushing ... the passers by' the mini buses crushed the passers by. 27b ani pl ha ... mi nibus ha aber ha ro krd /krd n mini bus pl passerby pl OM did.3SG /did 3PL the mini buses crushing ... the passers by' the mini buses crushed the passers by. 27c ani sg ... minibus ha aber ro krd /krd n mini bus pl passerby OM did.3SG /did 3PL the mini buses crushing ... the passer by' the mini buses crushed the passer by. 27d inani pl ha ... minibus ha motor ha ro krd /krd n mini bus pl motorcycle pl OM did.3SG /did 3PL the mini buses crushing ... the motorcycles' the mini buses crushed the motorcycle s. 27e inani sg ... minibus ha motor ro krd /krd n mini bus pl motorcycle OM did.3SG /did 3PL the mini buses crushing ... the motorcycle' the mini buses crushed the motorcycle.

PAGE 403

403 Table E 1. Continued item animacy NO Preamble gloss possible target response 28a ani pl an ... ruzname ha vrzeshkar an ro krd /krd n newspaper pl athlete pl OM approv did.3SG /did 3PL the newspapers approving ... the athletes' the newspapers approved the athletes. 28b ani pl ha ... ruzname ha vrzeshkar ha ro krd /krd n n ewspaper pl athlete pl OM did.3SG /did 3PL the newspapers approving ... the athletes' the newspapers approved the athletes. 28c ani sg ... ruzname ha vrzeshkar ro krd /krd n newspaper pl athlete OM did.3SG /did 3PL the newspapers approving ... the athlete' the newspapers approved the athlete. 28d inani pl ha ... ruzname ha snd ha ro krd /krd n newspaper pl document pl OM did.3SG /did 3PL the newspapers approving ... the documents' the newspapers approved the documents 28e inani sg ... ruzname ha snd ro krd /krd n newspaper pl document OM did.3SG /did 3PL the newspapers approving ... the document' the newspapers approved the document.

PAGE 404

404 Table E 1. Continued i tem animacy NO Preamble gloss possible target response 29a ani pl an ... ptu ha srbaz an ro krd /krd n blanket pl soldier pl OM did.3SG /did 3PL the blankets warming ... the soldiers' the blankets warmed the soldiers. 29b ani pl ha .. ptu ha srbaz ha ro krd /krd n blanket pl soldier pl OM did.3SG /did 3PL the blankets warming ... the soldiers' the blankets warmed the soldiers. 29c ani sg ... ptu ha srbaz ro krd /krd n blanket pl soldier OM did.3SG /did 3PL the blankets warming ... the soldier' the blankets warmed the soldier. 29d inani pl ha ... ptu ha txtexab ha ro krd /krd n blanket pl bed pl OM did.3SG /did 3PL the blankets warming ... the beds' the blankets warmed the beds. 29e inani sg ... ptu ha txtexab ro krd /krd n blanket pl bed OM did.3SG /did 3PL the blankets warming ... the bed' the blankets warmed the bed.

PAGE 405

405 Table E 1. Continued item animacy NO Preamble gloss possible target response 30a ani pl an ... hozche ha sb an ro krd /krd n pool pl horse pl OM did.3SG /did 3PL the pools disinfecting ... t he horses' the pools disinfected the horses. 30b ani pl ha ... hozche ha sb ha ro krd /krd n pool pl horse pl OM did.3SG /did 3PL the pools disinfecting ... the horses' the pools disinfected the horses. 30c ani sg ... hozche ha sb ro krd /krd n pool pl horse OM did.3SG /did 3PL the pools disinfecting ... the horse' the pools disinfected the horse. 30d inani pl ha ... hozche ha boshke ha ro krd /krd n pool pl barrel pl OM did.3SG /did 3PL the pools disinfecting ... the barrels' the pools disinfected the barrels. 30e inani sg ... hozche ha boshke ro krd /krd n pool pl barrel OM disinfecti did.3SG /did 3PL the pools disinfecting ... the barrel' the pools disinfected the barrel.

PAGE 406

406 Table E 1. Continued item animacy NO Preamble gloss possible target response 31a ani pl an ... tur ha mahi an ro krd /krd n net pl fish pl OM did.3SG /did 3PL the nets catching ... the fishes the nets caught the fishes. 31b ani pl ha ... tur ha mahi ha ro s krd /krd n net pl fish pl OM did.3SG /did 3PL the nets catching ... the fishes the nets caught the fishes. 31c ani sg ... tur ha mahi ro krd /krd n net pl fish OM did.3SG /did 3PL the nets catching ... the fish the nets caught the fish. 31d inani pl ha ... tur ha morvarid ha ro krd /krd n net pl pearl pl OM did.3SG /did 3PL the nets catching ... the pearls the nets caught the pearls. 31e inani sg ... tur ha morvarid ro krd /krd n net pl pearl OM did.3SG /did 3PL the nets catching ... the pearl' the nets caught the pearl.

PAGE 407

407 Table E 1. Continued item animacy NO Preamble gloss possible target response 32a ani pl an ... azmun ha davtlb an ro krd /krd n exam pl applicant pl OM did.3SG /did 3PL the exams testing ... the applicants' the exams tested the applicants. 32b ani pl ha ... azmun ha davtlb ha ro krd /krd n exam pl applicant pl OM did.3SG /did 3PL the exams testing ... the applicants' the exams tested the applicants. 32c ani sg ... azmun ha davtlb ro krd /krd n exam pl applicant OM did.3SG /did 3PL the exams testing ... the applicant' the exams tested the applicant. 32d inani pl ha ... azmun ha dstgah ha ro krd /krd n exam pl machine pl OM did.3SG /did 3PL the exams testing ... the machines' the exams tested the machines. 32e inani sg ... azmun ha dstgah ro krd /krd n exam pl machine OM did.3SG /did 3PL the exams testing ... the machine' the exams tested the machine.

PAGE 408

408 T able E 1. Continued item animacy NO Preamble gloss possible target response 33a ani pl an ... yaddasht ha moalem an ro krd /krd n note pl teacher pl OM did.3SG /did 3PL the notes describing ... the teachers' the notes described the teachers. 33b ani pl ha ... yaddasht ha moalem ha ro krd /krd n note pl teacher pl OM did.3SG /did 3PL the notes describing ... the teachers' the notes described the teachers. 33c ani sg ... yaddasht ha moalem ro krd /krd n note pl teacher OM did .3SG /did 3PL the notes describing ... the teacher' the notes described the teacher. 33d inani pl ha ... yaddasht ha mnzre ha ro krd /krd n note pl scene pl OM did.3SG /did 3PL the notes describing ... the scenes' the notes described the scenes. 33e inani sg ... yaddasht ha mnzre ro krd /krd n note pl scene OM did.3SG /did 3PL the notes describing ... the scene' the notes described the scene.

PAGE 409

409 Table E 1. Continued item animacy NO Preamble gloss possible target response 34a ani pl an ... jresqil ha kargr an ro bord /bord n crane pl labor pl OM took.3SG /took 3PL the cranes raised the labors. 34b ani pl ha ... jresqil ha kargr ha ro bord /bord n crane pl labor pl OM took.3SG /took 3PL the cranes raised the labors. 34c ani sg ... jresqil ha kargr ro bord /bord n crane pl labor OM took.3SG /took 3PL the cranes raised the labor. 34d inani pl ha ... jresqil ha snduq ha ro bord /bord n crane pl box pl OM took.3SG /took 3PL the cranes raised the boxes. 34e inani sg ... j resqil ha snduq ro bord /bord n crane pl box OM took.3SG /took 3PL the cranes raised the box.

PAGE 410

410 Table E 1. Continued item animacy NO Preamble gloss pos sible target response 35a ani pl an ... qfse ha dozd an ro krd /krd n shelf pl thief pl OM did.3SG /did 3PL the shelves hiding ... the thieves the shelves hid the thieves. 35b ani pl ha ... qfse ha dozd ha ro krd /krd n shelf pl thief pl OM did.3SG /did 3PL the shelves hiding ... the thieves the shelves hid the thieves. 35c ani sg ... qfse ha dozd ro krd /krd n shelf pl thief OM did.3SG /did 3PL the shelves hiding ... the thief' the shelves hid the thief. 35d inani pl ha ... qfse ha stl ha ro krd /krd n shelf pl bucket pl OM did.3SG /did 3PL the shelves hiding ... the buckets' the shelves hid the buckets. 35e inani sg ... qfse ha stl ro krd /krd n shelf pl bucket OM did.3SG /did 3PL the shelves hiding ... the bucket' the shelves hid the bucket.

PAGE 411

411 Table E 1. Continued item animacy NO Preamb le gloss possible target response 36a ani pl an ... helikupter ha gomshode gan ro krd /krd n helicopter pl lost one pl OM did.3SG /did 3PL the helicopters finding ... the lost people' the helicopters found the lost people. 36b ani pl ha ... helikupter ha gomshode ha ro krd /krd n helicopter pl lost one pl OM did.3SG /did 3PL the helicopters finding ... the lost people' the helicopters found the lost people. 36c ani sg ... helikupter ha gomshode ro krd /krd n helicopter pl lost one OM finding did.3SG /did 3PL the helicopters finding ... the lost person' the helicopters found the lost person. 36d inani pl ha ... helikupter ha jzire ha ro krd /krd n helicopter pl island pl OM did.3SG /did 3PL the helicopters finding ... the islands' the helicopters found the islands. 36e inani sg ... helikupter ha jzire ro krd /krd n helicopter pl island OM did.3SG /did 3PL the helicopters finding ... the island' the helicopters found the island.

PAGE 412

412 Table E 1. Continued item animacy NO Preamble gloss possible target respon se 37a ani pl an ... mashin ha zendani an ro dad /dad n car pl prisoner pl OM gave.3SG /gave 3PL the cars delivered the prisoners. 37b ani pl ha ... mashin ha zendani ha ro dad /dad n car pl prisoner pl OM gave.3SG /gave 3PL the cars d elivered the prisoners. 37c ani sg ... mashin ha zendani ro dad /dad n car pl prisoner OM gave.3SG /gave 3PL the cars delivered the prisoner. 37d inani pl ha ... mashin ha bste ha ro dad /dad n car pl parcel pl OM gave.3SG /gave 3PL the cars delivered the parcels. 37e inani sg ... mashin ha bste ro dad /dad n car pl parcel OM gave.3SG /gave 3PL l' the cars delivered the parcel.

PAGE 413

413 Table E 1. Continued item animacy NO Preamble gloss possible target response 38a ani pl an ... trazu ha qhrman an ro krd /krd n scale pl ch ampion pl OM did.3SG /did 3PL the scales weighing ... the champions' the scales weighed the champions. 38b ani pl ha ... trazu ha qhrman ha ro krd /krd n scale pl champion pl OM did.3SG /did 3PL the scales weighing ... the champions' the scales weighed the champions. 38c ani sg ... trazu ha qhrman ro krd /krd n scale pl champion OM did.3SG /did 3PL the scales weighing ... the champion' the scales weighed the champion. 38d inani pl ha ... trazu ha hendune ha ro krd /krd n scale pl watermelon pl OM did.3SG /did 3PL the scales weighing ... the watermelons' the scales weighed the watermelons. 38e inani sg ... trazu ha hendune ro krd /krd n scale pl watermelon OM did.3SG /did 3PL the scales weighing ... the watermelon' the scales weighed the watermelon.

PAGE 414

41 4 Table E 1. Continued i tem animacy NO Preamble gloss possible target response 39a ani pl an ... tank ha rzmnde gan ro krd /krd n tank pl soldier pl OM did.3SG /did 3PL the tanks chasing ... the soldiers the tanks chased the soldiers 39b ani pl ha ... tank ha rzmnde ha ro krd /krd n tank pl soldier pl OM did.3SG /did 3PL the tanks chasing ... the soldiers the tanks chased the soldiers. 39c ani sg ... tank ha rzmnde ro krd /krd n tank pl soldier OM did.3SG /did 3PL the tanks chasing ... the soldier the tanks chased the soldier. 3 9d inani pl ha ... tank ha rd e pa ha ro krd /krd n tank pl track pl OM did.3SG /did 3PL the tanks chasing ... the tracks the tanks chased the tracks. 39e inani sg ... tank ha rd e pa ro krd /krd n tank pl track OM did.3SG /did 3PL the tanks chasing ... the track the tanks chased the track.

PAGE 415

415 Ta ble E 1. Continued item animacy NO Preamble gloss possible target response 40a ani pl an ... qrardad ha davr an ro krd /krd n agreement pl referee pl OM did.3SG /did 3PL the agreements nominating ... the referees' the agreements nominated the referees. 40b ani pl ha ... qrardad ha davr ha ro krd /krd n agreement pl referee pl OM nominating did.3SG /did 3PL the agreements nominating ... the referees' the agreements nominated the referees. 40c ani sg ... qrardad ha davr ro krd /krd n agreement pl refer ee OM did.3SG /did 3PL the agreements nominating ... the referee' the agreements nominated the referee. 40d inani pl ha ... qrardad ha qeimt ha ro krd /krd n agreement pl price pl OM did.3SG /did 3PL the agreements nominating ... the prices' the agreements nominated the prices. 40e inani Sg ... qrardad ha qeimt ro krd /krd n agreement pl price OM did.3SG /did 3PL the agreements nominating ... the price' the agreements nominated the price.

PAGE 416

416 Table E 2. List of the filler items for Attraction experiment item P reamble gloss Possible response 1 .... reza v mnizheh bche ha ro krd n Reza and Manizheh child PL OM did 3PL Reza and Manizheh amusing ..... the children' Reza and Manizheh amused the children. 2 .... ehsan v omid baqch ha ro dad n Ehsan and Omid garden PL OM gave 3PL Ehsan and Omid watering ..... the gardens' Ehsan and Omid watered the gardens. 3 .... sudabeh v nstrn ayene ha ro krd n Sudabeh and Nastaran mir r or PL OM did 3PL Sudabeh and Nastaran cleaning ..... the mirrors' Sudabeh and Nastaran cleaned the mirr ors. 4 .... saman v hsn drs ha ro krd n Saman and Hasan lesson PL OM did 3PL Saman and Hasan reviewing ..... the lessons' Saman and Hasan reviewed the les sons. 5 .... mehri v sima ks ha ro krd n Mehri and Sima photo PL OM did 3PL Mehri and Sima looking ..... at the photos' Mehri and Sima looked at the photos 6 .... shirin v kaveh divar ha ro krd n Shirin and Kaveh wall PL OM did 3PL Shirin and Kaveh painting ..... the walls' Shirin and Kaveh painted the walls.

PAGE 417

417 Table E 2. Continued item preamble gloss Possible response 7 .... mina v susn kfsh ha ro zd n Mina and Susan shoe PL OM stroke 3PL Mina and Susan ..... waxes the sho es' Mina and Susan shined the shoes. 8 .... mn v kiyanush riyazi ro dasht im I and Kiyanush math OM had 1PL Kianush and I loving ..... Mathematics' Kianush and I loved Mathematics. 9 .... shbnm v mn she'r ro krd im Shabnam and I poem OM memorizing did 1PL Shabnam and I memorizing ..... the poem' Shabnam and I memorized the poem. 10 .... mn v susn film ro krd im I and Susan film OM did.1PL Susan and I watching ..... the movie' Susan and I watched the mo vie. 11 .... shkan v mn qza ro krd im Ashkan and I food OM did 1PL Ashkan and I preparing ..... the food' Ashkan and I prepared the food. 12 .... mn v krd im I and did.1PL Hadi and I swimming ..... Hadi and I swam.

PAGE 418

418 Table E 2. Continued item preamble gloss Possible response 13 ... s'id v to goldan ro krd in Saeed and you pot OM did.2PL Saeed and you painting ..... the pot' Saeed and you painted the pot. 14 .... mrym v nima nmayesh ro krd n Maryam and Nima play OM practicing did 3PL Maryam and Nima practicing ..... the play' Maryam and Nima practiced the play. 15 .... hosein bste ha ro krd Hosein parcel PL OM did.3SG Hosein weighing ..... the parcels' Hosein weighed the parcels. 16 .... roftegr xiyabun ha ro krd sweeper street PL OM did.3SG the street sweeper sweeping ..... the streets' the street sweeper swept the streets. 17 .... min klm ha ro krd Amin cab b age PL OM chopping did.3SG Amin chopping ..... the cab b ages' Amin chopped the cab b ages. 18 .... sina ms'le ha ro krd Sina problem PL OM did.3SG Sina solving ..... the problems' Sina solved the problems.

PAGE 419

419 T able E 2. Continued item preamble gloss Possible response 19 .... babk gusht ha ro krd Babak beef PL OM did.3SG Babak frying ..... the beefs' Babak fried the beefs. 20 .... sheida komod ha ro krd Sheida dresser PL OM did.3SG Sheida organizing ..... the dressers' Sheida organized the dressers. 21 .... li derxt ha ro krd Ali tree PL OM did.3SG Ali sawing ..... the trees' Ali saw the trees. 22 .... bita sho'le ha ro krd Bita flame PL OM did.3SG Bita putting out ..... the flames Bita put out the flames. 23 .... minu otaq ro krd Minu room OM vacuuming did.3SG Minu vacuuming ..... the room' Minu vacuumed the room. 24 .... j'fr stl ro krd Jafar bucket OM did.3SG jafar emptying ..... the buckets' jafar emptied the buckets.

PAGE 420

420 Table E 2. Continued item preamble gloss Possible response 25 .... giti brge ha ro mngene krd Giti paper PL OM stapli did.3SG Giti stapling ..... the papers' Giti stapled the papers. 26 .... mehran mehman an ro krd Mehran guest PL OM did.3SG Mehrane inv iting ..... the guests' Mehrane invited the guests. 27 .... rozita shm ha ro krd Rozita candle PL OM did.3SG Rozita turning on.... the candles' Rozita turned on the ca ndles. 28 seilab hme chiz ro ba xod sh krd flood every thing OM with itself it did.3SG the flood carrying ..... every thing with itself' the f lood carried every thing with itself. 29 .... baran hva ro krd rain weather OM did.3SG the rain purifying ..... the weather' the rain purified the weather. 30 .... qfse ye ketab ha kenar e bud shelf EZ book ha beside EZ was.3SG the book shelf ..... beside the table' the book shelf was beside the table.

PAGE 421

421 Table E 2. Continued item preamble gloss Possib le response 31 .... sib e qermez bu ye xub i dasht apple EZ red smell EZ good a had.3SG Red apple ..... a good smell' Red apple had a good smell. 32 yadgiri e zban e chini sxt bud learning language EZ Chinese difficult was.3SG Learning Chinese ..... difficult' Learning Chinese was difficult. 33 .... hva ye pa'izi kmi srd bud weather EZ fal l a little cold was.3SG Fall weather ..... a little cold' Fall weather was a little cold. 34 .... brg e narenji ruye zmin bud /oftad leaf EZ orange on ground was.3 SG /fall.3SG The orange leaf ..... on the ground' The orange leaf was on the ground. 35 .... film e ruz e jom'e didni bud film EZ day EZ Friday worth seeing was.3SG The film of Frid The film shown on Friday was worth seeing. 36 zrfshu'i kenar e pnjre bud sink beside EZ window was.3SG The sink ..... beside the window' The sink was beside the window.

PAGE 422

422 Table E 2. Continued item preamble gloss Possible response 37 .... aparteman e mn mnzre xub i dasht ap artment EZ I view good a had.3SG My ap artment ..... good view' My a partment h ad a good view. 38 .... salad e kahu tuye yxchal bud salad lettuce in fridge was.3SG the lettuce salad ..... in the fridge' the lettuce salad was in the fridge. 39 .... hva bri bud weather was.3SG the weather ..... cloudy' the weather was cloudy. 40 .... pakt e mive ruye miz e ashpzxune bud bag EZ fr uit on table EZ kitchen was.3SG the fruit bag ..... on the kitchen table' the fruit bag was on the kitchen table.

PAGE 423

423 LIST OF REFERENCES Akmajian, A., & Heny, F. (1975). An introduction to the prin ciples of transformational syntax Cambridge, MA: MIT press. Antn Mndez, I., Nicol, J. L., & Garrett, M. F. (2002). The relation between gender and number agreement processing. Syntax, 5 (1), 1 25. Baayen, R. H. (2008). Analyzing linguistic data: A pract ical introduction to statistics using R. Cambridge: Cambridge University Press. Baltin, M., & Collins, C. (2003). The handbook of contemporary syntactic theory Malden, Mass: Wiley Blackwell (pages 483 511). Barker, J., Nicol, J., & Garrett, M. (2001). Sem antic factors in the production of number agreement. Journal of Psycholinguistic Research, 30 (1), 91 114. Barlow, M. (1988). Unification and agreement. Paper presented at the Center for the study of Language and Information. Report NO. CSLI 88 120. Bar low, M. (1993). A suited theory of agreement, Garland Outstanding Dissertations in Linguistcs. Garland, New York. Bashiri, I. (1981). Persian Syntax Minneapolis: Burgess Publishing Company Bates, E., & MacWinney, B. (1989). Functionalism and the compet ition model. In B. MacWhinney & E. Bates (Eds.), The crosslinguistic study of sentence processing (pp. 10 75). Cambridge, England: Cambridge University Press. Bates, E., McNew, S., MacWhinney, B., Devescovi, A., & Smith, S. (1982). Functional constraints o n sentence processing: a cross linguistic study. Cognition, 11 245 299. Binnick, R. I. (1991). Time and the verb: A guide to tense and aspect Oxford University Press on Demand. Bock, K. (1991). A sketchbook of production problems. Journal of Psycholing uistic Research 20 (3), 141 160. Bock, J. K., Eberhard, K. M., & Cutting, J. C. (1992). Controlling number agreement on verbs and anaphors Bock, K. (1995). Producing agreement. Current Directions in Psychological Science, 4 (2), 56 61. Bock, K. (1996). L anguage production: Methods and methodologies. Psychonomic Bulletin & Review, 3 (4), 395 421.

PAGE 424

424 Bock, K., Butterfield, S., Cutler, A., Cutting, J. C., Eberhard, K. M., & Humphreys, K. R. (2006). Number agreement in British and American English: Disagreeing t o agree collectively. LANGUAGE BALTIMORE LINGUISTIC SOCIETY OF AMERICA 82 (1), 64. Bock, K., & Cutting, J. C. (1992). Regulating mental energy: Performance units in language production* 1. Journal of Memory and Language, 31 (1), 99 127. Bock, K., & Eberh ard, K. M. (1993). Meaning, sound and syntax in English number agreement. Language and Cognitive Processes, 8 (1), 57 99. Bock, K., Eberhard, K. M., & Cutting, J. C. (2004). Producing number agreement: How pronouns equal verbs. Journal of Memory and Langua ge, 51 (2), 251 278. Bock, K., Eberhard, K. M., Cutting, J. C., Meyer, A. S., & Schriefers, H. (2001). Some Attractions of Verb Agreement. Cognitive Psychology, 43 (2), 83 128. Bock, K., & Levelt, W. (1994). Language production: grammatical encoding In M. A. Gernsbacher (Ed.), Handbook of Psycholinguistics (pp. 945 984). San Diego: Academic press. Bock, K., & Miller, C. A. (1991). Broken agreement. Cognitive Psychology, 23 (1), 45 93. Bock, K., Nicol, J., & Cutting, J. C. (1999). The ties that bind: Creatin g number agreement in speech. Journal of Memory and Language, 40 330 346. Bock, K., & Middleton, E. L. (2011). Reaching agreement. Natural Language & Linguistic Theory 29 (4), 1033 1069. Boland, J. E. (2004). Linking eye movements to sentence comprehens ion in reading and listening. In M. Carreiras & C. Clifton Jr. (Eds.), The on line study of sentence comprehension: Eyetracking, ERP, and beyond (pp. 51 76). New York: Psychology Press. Boland, J. E., & Cutler, A. (1996). Interaction with autonomy: Multipl e output models and the inadequacy of the Great Divide. Cognition, 58 (3), 309 320. Carnie, A. (2007). Syntax: a generative introduction (2me d.): Oxford: Basil Blackwell. Chomsky, N. (1965). Aspects of the Theory of Syntax Cambridge, MA: MIT press. Clifton, C., Frazier, L., & Deevy, P. (1999). Feature manipulation in sentence comprehension. Rivista di linguistica, 11 11 39. Clifton, C., Staub, A., & Rayner, K. (2007). Eye movements in reading words and sentences. Eye movements: A window on mind and brain 341 372. Corbett, G. G. (2000). Number Cambridge: Cambridge University Press (pages 54 88).

PAGE 425

425 Corbett, G.G. (1979). The agreement hierarchy. Journal of Linguistics, 15, 203 224. Corbett, G.G. (1983). Hierarchies, Target and controllers: Agreement patterns in Slavic. University Park, PAl: Pennsylvania State University Press. Corbet, G. G. (1994). Agreement. In R. Asher, & J. Simpson Eds.), Encyclopedia of language and linguistics New York:Pergamon Press Dabir Moghaddam, M. (1997). Compound verb s in Persian. Studies in the Linguistic Sciences 27 (2), 25 59. Dabir Moghaddam, M. (1982). Passive in Persian. Studies in the Linguistic Sciences 12 (1), 63 90. Deutsch, A. (1998). Subject predicate agreement in Hebrew: Interrelations with semantic proc esses. Language and Cognitive Processes, 13 (5), 575 597. Deutsch, A., & Bentin, S. (2001). Syntactic and Semantic Factors in Processing Gender Agreement in Hebrew: Evidence from ERPs and Eye Movements* 1. Journal of Memory and Language, 45 (2), 200 224. D uffield, C. J. (2012). Subject verb agreement in English relative clauses: Using speech errors and psycholinguistic approaches to distinguish between syntactic representations. Working Papers of the Linguistics Circle 21 (2), 91 99. Eberhard, K. M. (1997). The Marked Effect of Number on Subject Verb Agreement. Journal of Memory and Language, 36 (2), 147 164. Eberhard, K. M. (1999). The accessibility of conceptual number to the processes of subject verb agreement in English. Journal of Memory and Language, 4 1 560 578. Eberhard, K. M., Cutting, J. C., & Bock, K. (2005). Making Syntax of Sense: Number Agreement in Sentence Production. Psychological Review, 112 (3), 531 559. Eberhard, K. M., Spivey Knowlton, M. J., Sedivy, J. C., & Tanenhaus, M. K. (1995). Eye movements as a window into real time spoken language comprehension in natural contexts. Journal of Psycholinguistic Research, 24 (6), 409 436. Fodor, J. A. (1983). The modularity of mind MIT press. Franck, J., Lassi, G., Frauenfelder, U. H., & Rizzi, L. (2006). Agreement and movement: A syntactic analysis of attraction. Cognition, 101 (1), 173 216. Franck, J., Soare, G., Frauenfelder, U. H., & Rizzi, L. (2010). Object interference in subject verb agreement: The role of intermediate traces of movement. Jo urnal of Memory and Language, 62 166 182.

PAGE 426

426 Franck, J., Vigliocco, G., & Nicol, J. (2002). Subject verb agreement errors in French and English: The role of syntactic hierarchy. Language and Cognitive Processes, 17 (4), 371 404. Fraser, N. M., & Corbett, G. G. (1994). Gender, animacy, and declensional class assignment: a unified account for Russian. In G. Booiij & J. Marle (Eds.), Yearbook of morphology (pp. 123 150). Dordrecht: Kluwer. Garnham, A., Shillcock, R. C., Brown, G. D., Mill, A. I., & Cutler, A. ( 1981). Slips of the tongue in the London Lund corpus of spontaneous conversation. Linguistics 19 (7 8), 805 818. Garnsey, S. M. (1993). Event related brain potentials in the study of language: An introduction. Language and Cognitive Processes, 8 (4), 337 3 56. Garrett, M. F. (1980). Levels of processing in sentence production. In B. Butterworth (Ed.), Language production, Volume 1: Speech and talk London: Academic Press. Gazdar, G., Kleen, E., Pullum, G., & Sag, I. (1985). Genneralized phrase structure gra mmar Oxford: Blackwell. Ghaniabadi, S. (2009). Definiteness marking through number Paper presented at the Annual conference of the Canadian Linguistic Association, University of Torento. Ghomeshi, J. (1996). Projection and inflection: A study of Persian phrase structure. PhD, University of Toronto. Hartsuiker, R. J., Antn Mndez, I., & van Zee, M. (2001). Object Attraction in Subject Verb Agreement Construction. Journal of Memory and Language, 45 (4), 546 572. Hartsuiker, R. J., & Barkhuysen, P. N. (2 006). Language production and working memory: The case of subject verb agreement. Language and Cognitive Processes, 21 (1), 181 204. Hartsuiker, R. J., Schriefers, H. J., Bock, K., & Kikstra, G. M. (2003). Morphophonological influences on the construction of subject verb agreement. Memory & cognition, 31 (8), 1316. Haskell, T. R., & MacDonald, M. C. (2003). Conflicting cues and competition in subject verb agreement. Journal of Memory and Language, 48 (4), 760 778. Haskell, T. R., & MacDonald, M. C. (2005). Constituent Structure and Linear Order in Language Production: Evidence From Subject Verb Agreement. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31 (5), 891 904.

PAGE 427

427 Hussler, J., & Bader, M. (2009). Agreement checking and number attra ction in sentence comprehension: Insights from German relative clauses. Travaux du cercle linguistique de Prague, 7 Humphreys, K. R., & Bock, K. (2005). Notional number agreement in English. Psychonomic bulletin & review, 12 (4), 689 695. Kaan, E. (2002) Investigating the effects of distance and number interference in processing subject verb dependencies: An ERP study. Journal of Psycholinguistic Research, 31 (2), 165 193. Karimi, S. (1989). Aspects of Persian syntax, specificity, and the theory of gramm ar. University of Washington. Karimi, S. (1990). Obliqueness, specificity, and discourse functions: R in Persian. Linguistic Analysis, 20 139 191. Kempen, G., & Hoenkamp, E. (1987). Incremental prodedural grammar for sentence formulation. Cognitive S cience, 11 201 258. Largy, P., & Fayol, M. (2001). Oral cues improve subject verb agreement in written French. International Journal of Psychology, 36 (2), 121 131. Levelt, W. J. M. (1989). From intention to articulation Cambridge, MA: MIT Press. Levin, B. (1993). English verb classes and alternations: A preliminary investigation Chicago: University of Chicago press. Lotfi, A. R. (2006). Agreement in Persian. Linguistik online, 29 MacDonald, M. C., Pearlmutter, N. J., & Seidenberg, M. S. (1994). Lexi cal nature of syntactic ambiguity resolution. Psychological review 101 (4), 676 703. MacWhinney, B., Bates, E., & Kliegl, R. (1984). Cue validity and sentence interpretation in English, German, and Italian. Journal of Verbal Learning and Verbal Behavior, 23 (2), 127 150. Mahootian, S., & Gebhardt, L. (1997). Persian New York: Routledge (as a reference). Mallinson, G., & Blake, B. J. (1981). Language typology: Cross linguistic studies in syntax (Vol. 46). Amsterdam: North Holland. Meshkat al dini, M. (198 7). An introduction to Persian transformational syntax Mashhad: Ferdowsi University Press.

PAGE 428

428 Mitchell, D. C. (2004). On line methods in language processing: Introduction and historical review. In M. Carreiras & C. Clifton Jr. (Eds.), The on line study of s entence comprehension: Eyetracking, ERPs and beyond (pp. 15 32). New York: Psychology Press. Mohammad, J., & Karimi, S. (1992). 'Light' verbs are taking over: Complex verbs in Persian. Proceedings of The Western Conference on Linguistics (WECOL) 5.195 212. Nicol, J. L. (1995). Effects of clausal structure on subject verb agreement errors. Journal of Psycholinguistic Research, 24 (6), 507 516. Nicol, J. L., Forster, K. I., & Veres, C. (1997). Subject Verb Agreement Processes in Comprehension. Journal of Memo ry and Language, 36 (4), 569 587. Pearlmutter, N. J. (2000). Linear versus hierarchical agreement feature processing in comprehension. Journal of Psycholinguistic Research, 29 (1), 89 98. Pearlmutter, N. J., Garnsey, S. M., & Bock, K. (1999). Agreement pro cesses in sentence comprehension. Journal of Memory and Language, 41 427 456. Perlmutter, D. M. (1978). Impersonal passives and the Unaccusative Hypothesis. Proceedings of the 4th Annual Meeting, Berkeley Linguistics Society, 157 89. Pollard, C., & Sag, I. A. (1988). An information based theory of agreement. : Standford University, Center for the Study of Language and Information (Report No. CSLI 88 132) Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research. Psychol ogical bulletin, 124 (3), 372 422. Rayner, K., & Sereno, S. C. (1994). Eye movements in reading: Psycholinguistic studies. In M. Gemsbacher (Ed.), Handbook of psycholinguistics (pp. 57 81). New York: Academic Press. Rayner, K., Sereno, S. C., Morris, R. K. Schmauder, A. R., & Clifton, C. (1989). Eye movements and on line language comprehension processes. Language and Cognitive Processes, 4 (3), 21 49. Richardson, D. C., & Spivey, M. J. (2004). Eye tracking: Characteristics and methods. In G. Wnek & G. Bowl in (Eds.), Encyclopedia of biomaterials and biomedical engineering (pp. 568 600). New York: Dekker. Saadat, E. (1996). Persian Language: Distinguishing between animate and inanimate Nashr e Danesh, 45 (86), 239 265. Sedighi, A. (2005). Animacy: the ove rlooked feature in persian. Paper presented at the Annual conference of the Canadian Linguistic Association, Ottawa.

PAGE 429

429 Sedighi, A. (2007). Agreement r estriction in Persian Portland, Oregon: Rosenberg. conceptual functional account of subject verb agreement in Persian. Language Sciences, 29 (6), 787 803. Spivey, M. J., Tanenhaus, M. K., Eberhard, K. M., & Sedivy, J. C. (2002). Eye movements and spoken language comprehension: Effects of visual context on syntactic ambiguity resolution* 1. Cognitive Psychology, 45 (4), 447 481. Staub, A. (2009). On the interpretation of the number attraction effect: Response time evidence. Journal of Memory and Language, 60 (2), 308 347. Staub, A. (2010). Response time dis tributional evidence for distinct varieties of number attraction. Cognition, 114 447 454. Stemberger, J. P. (1985). An interactive activation model of language production. Progress in the psychology of language 1 143 186. Tabor, W., & Tanenhaus, M. K. (1999). Dynamical models of sentence processing. Cognitive Science, 23 (4), 491 515. Taleghani, A. H. (2008). Modality, aspect and negation in Persian Amesterdam: John Benjamins Pub Co. Tanenhaus, M. K., Spivey Knowlton, M. J., Eberhard, K. M., & Sedivy, J. C. (1995). Integration of visual and linguistic information in spoken language comprehension. Science, 268 (5217), 1632. Tanenhaus, M. K., & Trueswell, J. C. (1995). Sentence comprehension. In J. MIller & P. Eimas (Eds.), Handbook of perception and cog nition: speech, language and cognition (pp. 217 262). San Diego, CA: Academic Press. Thacktson, J., & Wheeler, M. (1983). An introduction to Persian Cambridge, MA: Harvard University. Thornton, R., & MacDonald, M. C. (2003). Plausibility and grammatical a greement. Journal of Memory and Language, 48 (4), 740 759. Trueswell, J. C., Tanenhaus, M. K., & Garnsey, S. M. (1994). Semantic influences on parsing: Use of thematic role information in syntactic ambiguity resolution. Journal of Memory and Language, 33 285 318. Vigliocco, G. (1996). One or more labels on the bottles? Notional concord in Dutch and French. Language and Cognitive Processes, 11 (4), 407 442. Vigliocco, G., Butterworth, B., & Garrett, M. F. (1996a). Subject verb agreement in Spanish and Engl ish: Differences in the role of conceptual constraints. Cognition, 61 (3), 261 298.

PAGE 430

430 Vigliocco, G., Butterworth, B., & Semenza, C. (1995). Constructing subject verb agreement in speech: The role of semantic and morphological factors. Journal of Memory and L anguage, 34 (2), 186 215. Vigliocco, G., & Franck, J. (1999). When sex and syntax go hand in hand: Gender agreement in language production. Journal of Memory and Language, 40 455 478. Vigliocco, G., & Franck, J. (2001). When sex affects syntax: Contextua l influences in sentence production. Journal of Memory and Language, 45 (3), 368 390. Vigliocco, G., & Hartsuiker, R. J. (2002). The interplay of meaning, sound, and syntax in sentence production. Psychological bulletin, 128 (3), 442 472. Vigliocco, G., Ha rtsuiker, R. J., Jarema, G., & Kolk, H. H. J. (1996b). How many labels on the bottles? Notional concord in Dutch and French. Language and Cognitive Processes, 11 (3), 407 421. Vigliocco, G., & Kita, S. (2006). Language specific properties of the lexicon: I mplications for learning and processing. Language and Cognitive Processes, 21 (7), 790 816. Vigliocco, G., & Nicol, J. (1998). Separating hierarchical relations and word order in language production: is proximity concord syntactic or linear? Cognition, 68 ( 1), B13 B29. Vigliocco, G., & Zilli, T. (1999). Syntactic accuracy in sentence production: The case of gender disagreement in Italian language impaired and unimpaired speakers. Journal of Psycholinguistic Research, 28 (6), 623 648. Wagers, M. W., Lau, E. F., & Phillips, C. (2009). Agreement attraction in comprehension: representations and processes. Journal of Memory and Language, 61 (2), 206 237. Wertheimer, M. (1993). Untersuchungen zur Lehre von der Gestalt, II. Psychologische Forschung, 4 301 350. Wi ndfuhr, G. (1979). Persian grammar: History and state of its study : Mouton De Gruyter.

PAGE 431

431 BIOGRAPHICAL SKETCH Aazamosadat Feizmohammadpour received a BA in English Translation from Iran and an MA in General Linguistics from Tarbiyat Modarres University, Tehran Iran. For her MA thesis, she conducted a sociolinguistic study of English language textbooks used in Iranian classrooms. She has studied Ph.D of Linguistics at the University of Florida. For her dissertation, she has conducted five experiments that test the effect of different factors on optional subject verb number agreement processing in Persian. Her primary research interest is in the intersection of the structure of Persian and psycholinguistics.