Title Page
 Table of Contents
 List of Tables
 Previous relevant Spanish phonological...
 Some phonological processes from...
 Some theoretical implications from...
 Summary and conclusions
 Biographical sketch

Title: Some theoretical implications from rapid speech phenomena in Miami-Cuban Spanish /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00097504/00001
 Material Information
Title: Some theoretical implications from rapid speech phenomena in Miami-Cuban Spanish /
Physical Description: xiii, 277 leaves : ; 28cm.
Language: English
Creator: Hammond, Robert Matthew
Donor: unknown ( endowment )
Publication Date: 1976
Copyright Date: 1976
Subject: Spanish language -- Dialects -- Florida -- Miami   ( lcsh )
Linguistics thesis Ph. D   ( lcsh )
Dissertations, Academic -- Linguistics -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 268-275.
Additional Physical Form: Also available on World Wide Web
Statement of Responsibility: by Robert M. Hammond.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00097504
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000169487
oclc - 02905063
notis - AAT5895

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
        Page vii
    List of Tables
        Page viii
        Page ix
        Page x
        Page xi
        Page xii
        Page xiii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
    Previous relevant Spanish phonological studies
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
    Some phonological processes from Miami-Cuban Spanish
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
    Some theoretical implications from rapid speech phenomena
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
        Page 245
        Page 246
        Page 247
        Page 248
        Page 249
        Page 250
        Page 251
        Page 252
        Page 253
        Page 254
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
    Summary and conclusions
        Page 262
        Page 263
        Page 264
        Page 265
        Page 266
        Page 267
        Page 268
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
        Page 275
    Biographical sketch
        Page 276
        Page 277
        Page 278
        Page 279
        Page 280
        Page 281
Full Text



Robert M. Hammond




Robert M. Hammond


This dissertation is dedicated to the numerous persons who

have been helpful during its preparation, especially the native

speakers of Miami-Cuban Spanish who served as informants. It is

likewise dedicated to those excellent professors whom I was fortunate

enough to have studied under, who have immeasurably contributed to my

understanding of language and linguistics, and also to the many other

colleagues who have indirectly been helpful during the preparation of

this study.


I would like to thank the members of my supervisory committee,

Professors Chauncey Chu, Gary Miller, Irving Wershow, and Manuel

Carvajal for their help and guidance during the preparation of this

dissertation, I would especially like to acknowledge my gratitude to

the director of my supervisory committee, Professor Bohdan Saciuk, who,

during my years of study at the University of Florida, has been the

single most positive influence on my academic life through his example

as an excellent teacher, counselor, researcher and scholar. A debt of

gratitude, of course, is owed to the many native speakers of MCS, too

numerous to mention, who have served as informants for this study.

Also, thanks are due to many of my former professors and colleagues

who have given me advice and ideas that have greatly added to the depth

of my work. I would also like to thank Mrs. Sylvia Fitzraurice for

typing this rather lengthy study. Finally, a debt of gratitude is

owed to Maggie MacDonald, fellow linguist and friend, for many hours

of help in the preparation of this study.








1.1 Purposes and Objectives
1.2 The Miami-Cuban Spanish Dialect Area
1.3 Methodology
1.4 A Definition of Rapid Speech


2.1 Spanish Generative Phonology
2.2 Cuban Spanish
2.2.1 Published materials on Cuban Spanish

Almendros (1958)
Barto (1965
Bartos (1966)
Bartos (1970)
Espinosa (1935)
Haden and Iatluck (1973 and 1974)
Isbagescu (1968a)
Isbasescu (1968b)
L6pez Morales (1970)
Terrell (1975a
Terrell (1975b
Terrell 1975c
Terrell 1975d
Terrell 1976a
Other Published Materials on Cuban Spanish

2.2.2 M.A. and Ph.D. Theses on Cuban Spanish

Bertot (1969)
Clegg (1967)
Guitart (1973)
Lamb (1968)
Sosa (1974)
Vallejo-Claros (1970)

2.2.3 Papers Presented on Cuban Spanish

viii Guitart (1974) 72
2.2.4 Unpublished Studies on Cuban Spanish 74 Terrell (1974a) 75 Terrell (1974b 76
2.3 Miami-Cuban Spanish 79
2.3.1 Published Katerials on :CS 79
2.3.2 Masters and Doctoral Research Done on YCS 80
2.3.3 Papers Read on IMCS 82
2.4 Other Spanish Dialect Areas 83
2.4.1 Cassano (1972) 84
2.4.2 del Rosario (1956) 84
2.4.3 Honsa (1965) 85
2.4.4 Hooper (1973) 86
2.4.5 Vondejar (1970) 87
2.4.6 Navarro-Tomns (1957) 88
2.4.7 Navarro-Toas (1966a) 89
2.4.8 Navarro-Tomas 1966b) 90
2.4.9 Tessen (1974) 90

3.1 Introduction 94
3.2 Methodology 96
3.3 The Velar Nasal in NCS Rapid Speech 105
3.4 Voicing of Intervocalic Stops 127
3.5 Phonetic Manifestations of Systematic /s/ 130
3.6 Final Consonant Deletion 141
3.6.1 /h/ in Word-final Position 141
3.6.2 /d/ in Word-final Environments 142
3.7 Phonetic Realizations of Systematic /c/ 143
3.8 Devoicing of Word-final Vocalic Segments 146
3.9 Monophthongization 147
3.10 Surface Ianifestations of Liquids 149
3.10.1 Phonetic Realizations of Systematic // 149
3.10.2 Phonetic Realizations of Systematic // 152
3.10.3 Phonetic Realizations of Systematic/r/ 155
3.11 Some Iiscellaneous Phonological
Processes of I.:CS Rapid Speech 165
3.11.1 Voicing Assimilation 165
3.11.2 The Dorsovelar Spirant 166
3.11.3 The Voiceless Bilabial Spirant 166
3.11.4 Intervocalic /d/ Deletion 167
3.12 Conclusions 167


4.1 Global Rules in A Generative Grammar 172
4.2 The Domain of "Final e Apocope"--A
Derivational Constraint 173
4.3 Compensatory Vowel Lengthening--a Global Rule 185
4.4 A Proposal for a Theory of Explanation in
Phonology--a Translexical Constraint 194
.4.5 Regularization of Compounds 198
4.6 The Relationship of stridenti and [high]
to Surface Forms of Non-liquid Consonants 202
4.7 Relative Narkedness and [+nasal] Segments 209
4.8 Word-boundary Phenomena 222
4.9 Linguistic Change--Sound Change and Spread 227
4.9.1 The Velar Nasal 228
4.9.2 Voicing of Intervocalic Stops 233
4.9.3 Spirantization and Systematic Phonemic
Representation of Voiced Obstruents in MCS 234 Arguments in Favor of [-continuanti
Systematic Phonemes for Voiced Obstruents
in the Synchronic Gramnar of MCS 237 Arguments in Favor of [+continuanti
Systematic Phonemes for Voiced Obstruents
in the Synchronic Grammar of ECS 243 The Naturalness Argument--a Rule of
Spirantization vs. a Rule of
Despirantization 243 Summary and Conclusions--/BDG/ in MCS 244 Epilogue--/BDG/ in KCS 246
4.9.4 Syllable-initial s Aspiration 247
4.9.5 Final Consonant Deletion 248
4.9.6 The Occurrence of [si 250
4.10 Conclusions to Chapter Four 252

5.1 Introduction 262
5.2 Summary and Conclusions--Previous Research
Related to MCS 263
5.3 Summary and Conclusions-Rapid Speech
Rules and Data from KCS 263
5.4 Summary and Conclusions--Theoretical
Implications 265



































APPEARS AS [r] or [a']

APPEARS AS [e] or ['Z]

APPEARS AS [1] or [' ]

APPEARS AS [rt or [7?]

APPEARS AS [Z] or [6 ]








































FOLLOWING [ segment]







and [91 BASED ON






























syllabicc] SEGMETTS










Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy



Robert M. Hammond

June, 1976

Chairman Bohdan Saciuk
Major Departments Linguistics

This study has three distinct, yet interrelated sections. The

first of these presents an annotated bibliography of all previous

phonological studies done on Cuban Spanish and on the Miami-Cuban dia-

lect of Spanish. The second major section presents phonological rules

and data from Miami-Cuban Spanish rapid speech. The third major por-

tion discusses theoretical implications which are drawn from the data

presented in the previous two sections.

The annotated bibliography presented in this study gives a de-

tailed, critical review of all published books and articles as well as

Ph.D. and M.A. theses done on Cuban Spanish and on Miami-Cuban Spanish.

This section also includes all discoverable papers on these dialects

read at academic conferences, as well as several unpublished studies.

Also contained in this annotated bibliography are a number of phono-

logical studies of several other Spanish dialect areas which are of

interest to the present study.

The second major section of this study presents data concerning

fourteen different consonantal segments and one process pertaining to

vocalic segments in Miami-Cuban Spanish rapid speech. Percentages of

occurrence for each of these different segments, along with variable

rules stated in terms of percentages of application are also given.

There are two significant contributions to the study of dialectal pho-

nology provided in this sections (1) It deals specifically with rapid

speech; (2) It gives actual percentages of occurrence of the different

surface manifestations of different systematic phonemes of Miami-Cuban

Spanish as well as the percentages of application of each portion of these

variable rules which are needed to account for these surface outputs.

Based primarily on the rapid speech data presented in the previous

section, the final portion of this study discusses a series of theoretical

questions which surface as a result of these data. Some of the traditional

analyses of Spanish phonological processes found in various generative

descriptions of different Spanish dialects, e.g., the underlying C-continuant]

status of the voiced obstruents /b d g/, are questioned, and alternative

analyses are proposed. Also, some of the traditional analyses are rejected

because they depend on the use of extrinsic rule ordering, highly abstract

underlying representations, etc., notions which are rejected herein, as they

have been previously rejected by other linguists, for having little or no

psychological reality.

Based on the relatively small corpus of data under analysis in the

present study (approximately five hours of recorded social conversation),

it appears evident that many phonological processes that occur in Miami-

Cuban Spanish rapid speech are difficult to account for using the traditional

generative approach, or for that matter, any other heretofore proposed lin-

guistic theory, e.g. the surface manifestations of nasal segments in this

dialect. Such data, it would seem, force us to either reject, or at least


reanalyze our thinking relative to standard nasal assimilation, at least

for rapid speech. If a relatively small corpus of rapid sp-.3ch data from

a single Spanish dialect can cause a number of weaknesses to surface in

some of our current descriptions of Spanish, it seems very likely that a

greater amount of data on rapid speech, including those from other Spanish

dialects, may well have an impact on our current phonological model as well

as in our ultimate phonological description of Spanish, and also on those

of other languages as well.





1.1 Purposes and Objectives

Because of many complex historical, political, and socio-economic

factors, since 1492 the language of Spain has evolved into many diverse

dialects and sub-dialects in the New World. Even in 1492 the language

of Spain was far from being a homogeneous entity, but rather, a variety

of dialects, already stratified along different lines of political

and/or religious demarcations. The Spanish of Granada, for example,

was distinct from that of Madrid, Barcelona, or Bilbao.1 The stratifi-

cation of peninsular Spanish in 1492 and later has, in many ways,

contributed to the development of American-Spanish dialects as they

exist today, although the situation is apparently not as simple as has

been previously supposed.2 One probably would not want to deny that the

initial influx of settlers from Southern Spain (i.e. Andalucfa) into the

Caribbean area shortly after Columbus' first voyage has had some in-

fluence on the Spanish spoken in this area today. The determination of

the precise degree of importance of these early settlement patterns

will be left to others. Another important factor which has had an

impact on current dialect areas in the New World is the influence of the

different indigenous peoples inhabiting these different areas of America

from 1492 to date. The indigenous languages of America, e.g. Nghuatl in

Mexico, Guarani in Paraguay, Quechua in Bolivia and Peru, have had

various degrees of influence, albeit mostly lexical, on the dialects of

Spanish spoken in these areas.3 Besides the indigenous languages of



America, the importation of various non-Hispanic peoples and tech-

nologies has also had an influence, again mostly lexical, on

American-Spanish dialect areas.

Although all these dialects and sub-dialects are a synchronic

fact, very few have been described within any linguistic theory, and

even fewer of them within the framework of generative grammar.

With regard to Cuban Spanish in particular, there are several

partial dialect studies available, treating the speech of certain

areas, e.g. the province of La Habana, but no complete phonological de-

scription has ever found its way into print.5

The situation with regard to Miami-Cuban Spanish (henceforth MCS)

is even worse, since there is but a handful of studies ever done on


Part of the purpose of this study will be to provide a partial

phonological description of this dialect of Spanish spoken in the Miami,

Florida area. The phonological description provided in this study will

be only partial for two reasons: (1) it will describe only phonological

processes that occur in rapid speech; (2) it will describe within the

theory of generative grammar only some of the rapid speech phonological

processes that occur in MCS that do not occur in the standard language.

This is not to imply, however, that these phonological processes occur

exclusively in MCS, as some or all of these individual processes have

been reported in one or more other dialects of Spanish. These rapid

speech phenomena will be discussed in Chapter Three of this disserta-


The major contribution to Spanish Generative Phonology of

Chapter Three is not that of simply presenting these data, as some or

all of them are known to those doing research in Cuban Spanish, but

rather, that it will (1) present these rapid speech phenomena within

the framework of generative theory; (2) show them in terms of frequency

of occurrence (i.e. percentages of occurrence) rather than simply de-

scribing them as occurring "often, sometimes, frequently," etc. A

secondary contribution of this study will be that it will present these

rapid speech phonological processes, thereby providing a rather large

and detailed body of data that others, not thoroughly familiar with

MCS, may use for testing different aspects of universal phonological

theory. It should be made clear, however, that it is not the purpose

in the present study to provide a complete phonological description of

MCS, but rather, as stated above, to account for a variety of rapid

speech phonological phenomena that occur in this dialect that do not

occur in the standard language.

The second major contribution that this study will make to

Spanish generative phonology in particular, and to universal phonologi-

cal theory in general, is that of discussing some of the theoretical

implications that arise from these rapid speech phenomena. Obviously,

since there have been no prior phonological studies done on MCS within

the theoretical framework of generative grammar, there has been no re-

search done on the theoretical implications that arise from these late

phonological processes. These rapid speech phenomena will be discussed

in Chapter Four as they relate to some current issues in generative

phonological theory.

Chapter One of this study will discuss purposes and objectives,

defining the MCS dialect area and rapid speech as they are understood

herein, and will also discuss methodology used in the data gathering


Chapter Two will review, discuss, and provide a critical analysis

of all available studies done on Cuban Spanish, MCS, and various other

dialect areas that exhibit some of the phonological processes occurring

in MCS.

The final portion of this study, Chapter Five, will provide a

summary and conclusions relative to the material presented in previous

chapters, and will offer some suggestions for further work that should

be done on MCS phonology.

While this study deals only with certain synchronic phonological

facts of MCS, since there is very little published research available

on this dialect, there is an abundance of worthwhile projects that can

and should be done on 1CS. As mentioned above, further study could be

undertaken on diachronic phonology and dialectology of Cuban Spanish

and Miami-Cuban Spanish. There is a wealth of socio-linguistic re-

search that could be done on MCS and the Spanish of the Tampa, Florida

area. Tampa Spanish, which has a strong Cuban influence, is quite

different from the Cuban Spanish spoken in Miami.7 A comparison of

the lexicology, syntax, and phonology of Tampa and Miami Spanish would

appear to be a significant linguistic study. Also, in the Miami area,

there is a sizeable number of bilingual Chinese (Chinese/Spanish), many

of whom came to Havana after the communist rise to power in China.

Many of these Chinese moved on to Miami after Castro's takeover in

Cuba. These speakers would seem to provide a source of data for dif-

ferent types of linguistically significant studies. Besides phonologi-

cal studies, the MCS dialect provides a good source for lexical and

syntactic studies. There is an abundance of lexical material from MCS

and Cuban Spanish which differ from standard Spanish, e.g. the lexical

item ahorita does not mean 'now' or 'right now' as in standard Spanish,

but rather 'in a little while' in Cuban Spanish and in MCS. There are

also syntactic characteristics peculiar to Cuban Spanish and MCS, e.g.

iqug tu haces? 'What are you doing?' is an acceptable surface level

manifestation of standard Spanish AQu6 haces tu? In standard Spanish,

questions of the form'Qug t-6 haces? are ungrammatical.8 There are

also other ways in which the syntax of MCS differs from that of the

standard language, some of which could have an influence on universal

syntactic theory.

1.2 The Miami-Cuban Spanish Dialect Area

Throughout this study reference is made to the Miami-Cuban

Spanish dialect. Rather than being limited only to Miami, this dialect

encompasses the entire coast of southeast Florida. This southeastern

coastal area is mainly comprised of Dade County, in which Miami is the

principal city, Broward County, whose main city is Fort Lauderdale,

and Palm Beach County, with West Palm Beach being its largest city.

The vast majority of speakers of MCS, to be sure, reside in three main

areas of Dade County. One of these main areas is the near southwest

quadrant of the city of Miami. This section is known as the sauesera

Csagwesgral, a Spanish phonological adaptation of the English word

'southwest,' with the addition of the Spanish suffix -era. The section

of Miami known as "Little Havana" is found here in "la saguesera," on

and around Southwest Eighth Street.

Another principal area of concentration of speakers of MCS is the

city of Hialeah, a northwestern suburb immediately bordering on the

city of Miami.

The third area of concentration of speakers of MCS is the south-

western suburb of Westchester. There are, of course, other areas of

concentration of speakers of MCS in this three-county area, but none as

large or as culturally dominating as these three areas.

The vast majority of the speakers of MCS who reside in this

three-county area are, of course, recent immigrants, who have arrived

sometime after 1959. Most speakers of MCS are, to varying degrees,

bilingual. Some exhibit a nearly perfect command of both English and

Spanish, most have some degree of ability to function in English (some

of course with a good command of English, others with only a basic

knowledge of English), and some have managed to remain almost com-

pletely monolingual speakers of Spanish. Speakers of MCS represent all

possible professions, socio-economic levels, and educational levels. A

large portion of the adults, at this time, have received some or most

of their education in Cuba. Speakers of MCS come from all six Cuban

provinces, and they represent various lengths of residence in the Miami

area, some having been here for almost sixteen years, others having

arrived only recently.

1.3 Methodology

The dialect of Spanish used in this study is that of twenty-one

personal friends, associates, and colleagues of the writer. All of

the data used in this study have been collected and recorded since

February, 1975. All informants used in this study are native speakers

of Cuban Spanish, who have resided principally in the Miami, Florida

area since their arrival in this country, and who still use Spanish as

their principal language. They represent various stages of bilingual-

ism. They are all adult speakers, representing an age range of 20-55

years. Most are educated speakers, although many have received some or

most of their education here in the United States. They represent five

of the six Cuban provinces.9


The speech of the informants used in this study is believed to be

typical of the Spanish spoken in what has been designated as the MCS

dialect area.

It should be made clear, however, that while what is described in

this study is felt to be typical VCS, the claim is not that MCS is

Cuban Spanish. This may well be the case-this is an empirical

question which only further research can determine. The limited number

of informants used in this study (21) should also be kept in mind.

Definitive and sweeping generalizations made in several recent studies

done on Cuban Spanish, based on a small number of informants, merely

begs the question. Until it can be shown that the limited number of

informants used in any or all of these studies are truly representative

of the Spanish spoken in Cuba, then this type of assumption can be

little more than just that-an assumption.

Another point which should be made clear is that this study makes

no direct attempt to determine the homogenity of the MCS dialect. It

is intuitively felt, however, that the speech of the informants used in

this study is typical of the MCS dialect area. The determination of

the homogenity or stratification of the Kiami-Cuban dialect of Spanish,

albeit an important question, is beyond the scope of this study.

In collecting the data used in this study, only high-quality

cassette recording equipment and tapes were used.

Since all the phonological processes which we present in

Chapter Three, and the theoretical implications presented in Chapter

Four are from rapid speech, all of the recorded data are taken from

totally unstructured "free-speech" situations. These four

"free-speech" recording sessions, in which most of the informants are

personal friends of the writer, were done in a social atmosphere,

wherein normal, everyday conversation was taking place.


In determining the frequency of occurrence of each of the surface

level variants discussed in Chapter Three, it was necessary to make some

arbitrary decisions. In section 3.5, where the frequency of occurrence

of the three surface level manifestations of systematic /s/. i.e. [s],

[h], or []1 was being determined, arbitrary parameters had to be set up.

In this section, as well as in all other sections, these somewhat

arbitrary parameters were set up using what was considered, under the

circumstances, the best possible linguistic judgement. Once these para-

meters had been established, care was taken to be consistent in all

judgements, staying within these self-imposed parameters.

In the case of section 3.5, only + strident] sibilants were

considered as occurrences of Cs1; only those surface manifestations of

[s] that contained no spirant quality whatsoever were considered as

occurrences of [Cj. Whatever was neither [s] nor [~] had to be con-

sidered as phonetic occurrences [h] of systematic /s/.

Because of the arbitrary decisions involved in these frequency of

occurrence determinations, the figures for percentage of occurrence must

be considered as perhaps a first approximation. This is, as far as is

known, the first attempt that has ever been made to ascertain this type

of information. In some ways, because of the importance that these

percentages potentially have on universal phonological theory, it is

somewhat surprising that this type of research has never been previously

undertaken, but considering the rather deplorable state of our knowledge

of Spanish dialectology in general, nothing is really surprising. Only

when others have undertaken the task of determining these frequency of

occurrence percentages, thereby providing a larger source of data, will

we have a more accurate knowledge of their reality.

1.4 A Definition of Rapid Speech

In demarcating different styles of speech, it was again obliga-

tory to establish some more or less arbitrary parameters. It is ne-

cessary, nevertheless, to define these different styles of speech,

particularly what is meant when referring to "rapid speech," in order

to establish a framework far the discussion in Chapters Three and Four

of this study.

In the area of Spanish phonological studies, Harris (1969) was,

as far as can be determined, the first person to formalize different

styles of speech, to give "an impressionistic description of a hier-

archy of 'styles' of pronunciation in Spanish" (1969, p. 7). Harris

(1969) gives the following parameters for the four different styles of

speech he recognizes:

Laro: very slow, deliberate, overprecise; typical of,
for example, trying to communicate with a for-
eigner who has little competence in the language
or correcting a misunderstanding over a bad tele-
phone connection.

Andantes moderately slow, careful, but natural; typical
of, for example, delivering a lecture or teaching
a class in a large hall without electronic

Allegrettot moderately fast, casual, colloquial. In
many situations one might easily alternate be-
tween andante and allegretto in mid-discourse
or even in mid-sentence.

Prestot very fast, completely unguarded.

Other Spanish phonologists have, of course, been aware of how

different styles of speech affect surface phonetic forms. Navarro-

Tomas (1965) makes repeated references to different speeds and levels

of speech. Harris' treatment of these styles of speech is unique in

that he actually formalizes them, not that he is the first to recognize

the role they play in phonological processes.

Harris (1969, ad passim) suggests or implies that normal, rapid

speech is represented by the styles he calls allegretto and presto. He

suggests that "in many situations one might easily alternate between

andante and allegretto (1969, p. 7). It is suggested that, in normal

colloquial speech, it is perhaps more normal to alternate between

Allegretto and presto.

In establishing parameters for a definition of rapid speech,

Harris' definitions of his styles of allegretto and presto can almost,

but not quite, be accepted. The following definition is offered as

minimal parameters for what is referred to as "rapid speech" throughout

this study

RaPid Speechs normal, fast, casual, unguarded colloquial
speech, typical of any normal social situation.

This definition differs from Harris' styles of allegretto and

presto in that it does not accept the slowest area of the allegretto

style of speech, i.e. moderately fast. It might be said that the style

of speech which is defined as "rapid speech" encompasses the upper-half

of Harris' allegretto style, and of course presto. The parameters

given in this definition are, of course, only the minimal parameters,

i.e. it also allows speech faster than "normal"; it allows "completely

unguarded speech," etc.


1See Lapesa (1959), Menendez Pidal (1962), Entbstle (1938), etc.
2For example, see Almendros (1958) for an oversimplistic view.
Although the question of the precise influence that these peninsular
dialects may have had on current American-Spanish dialects is an in-
teresting and important one, it properly belongs in the realm of
diachronic Spanish dialectology, and therefore is not within the scope
of this dissertation.

3See Buesa Oliver (1965), Tessen (1974), Boyd-Bowman (1960), etc.

4See Boyd-Bowman (1960), Grayson (19&g), Hammond (1974),
ITavaez (1963), etc.

5rhese studies will be discussed in Chapter Two of this study.

6See Saciuk (1974), Hammond (1973), Resnick and Hammond (1973 and
1975), Hammond and Resnick (1974), Bjarkman (forthcoming), Fernandez
(1973), Bertot (1969). Also, Bohdan Saciuk has been working on an
extensive and detailed study of the phonology of Miami-Cuban Spanish.

7Personal observation of the writer of this study, who was raised
in the Tampa Spanish area.

8rhe example given here is not merely an isolated idiosyncratic
occurrence of this phenomenon. This is a rather productive process
with some verbs in the second-person singular form. Other common
examples are, L Ad6nde tu vas?, 'Where are you going?'; iQ u6 tu crees?,
'What do you think?'; bD6nde tu vives?, 'Where do you live?'; Agu6 tt
mirs?, 'What are you looking at?'. This process occurs primarily with
the subject pronoun tf, less frequently with the pronoun used or
ustedes, and from time to time with the first-person pronouns. I
cannot recall having ever heard this process with the other third-per-
son pronouns, with any degree of regularity (i.e. 61, ella, ellos,
ellas). The precise constraints which apply to this process, i.e. with
which verbs it may occur, with which WH-forms it may occur, and in what
styles of speech it may occur, etc., would have to be confronted in
such a study.

90ther background data relative to the twenty-one informants is
given in section 3.2 of this study.


2.1 Spanish Generative Phonology

Although Harris (1969) described a sub-dialect of Mexican Spanish,

it is perhaps appropriate to briefly discuss its impact on the present

study. While he was directly describing only one dialect of American

Spanish, he was also indirectly describing some phonological processes

in all other dialects of Spanish; e.g., the phonological process of

velarr softening' is shown informally in Harris (1969, p. 71)t

k -cons
g '- back

The effect of this rule is to change systematic /k/ and /g/ to Cs7 and

[x] respectively, before a front vowel.1 This process of velar soften-
ing occurs in this same way in numerous other dialects of Spanish.2

Thus, Harris (1969) also describes many phonological processes of MCS,

e.g., his rule of 'spirantization' (rule 6.5(36), p. 187) in Mexican

Spanish correctly describes this same process in MCS. It is therefore

no coincidence that the late rapid speech phonetic processes described

in Chapter Three of this study are processes not described in Harris

(1969). To attempt to repeat the rules found in Harris (1969) that

correctly describe phonological facts of MCS would be of little in-

terest or value.

This study uses Harris (1969) as a starting point for the phono-

logical processes described in Chapter Three, because of the belief that

it is the first complete, reasonably accurate phonological description

of any Spanish dialect ever described within the framework of generative


While it is not the claim of this study to be providing a complete

description of the phonology of MCS, if Halle (1962, p. 343) in discus-

sing dialectology is correct in saying that "differences among grammars

(i.e., dialects-R.M.H.] may be due to one or both of the followings

(a) different grammars may contain different rules; (b) different

grammars may have differently ordered rules," then it seems reasonable

to assume that by incorporating Harris' (1969) rules that correctly de-

scribe phonological facts of MCS, by deleting those rules of Harris

(1969) that do not apply to MCS, and by adding the phonological rules of

MCS found in Chapter Three of this study, we would have a fairly com-

plete phonological description of MCS. It perhaps should be clarified

at this point that there is no implication here to suggest that Mexican

Spanish, or any other dialect of Spanish be "converted" into MCS, i.e.,

that its phonetic representations be taken as the underlying representa-

tions of MCS, an approach espoused by some,4 but rather that MCS and

Mexican Spanish share some phonological rules, and that these rules

shared by MCS and Mexican Spanish described in Harris (1969), along with

the phonological rules found in Chapter Three of this study should pro-

vide a nearly complete phonological description of the phonological

component of MCS.

2.2 Cuban Spanish

Although linguistically, politically, and socio-economically Cuba

has been an important part of the Caribbean area for many years, there

is a relative paucity of studies done on the dialect of Spanish spoken

on the island. Because of obvious political reasons, there have not been

many dialect studies done in Cuba since 1959, but there remains a long

period of time, between 1493 and 1958, during which such studies could

have been carried out. The dearth of phonological studies done on Cuban

Spanish is indicative of the rather poor state of the study of

American-Spanish dialectology in general.

Of the studies done on Cuban Spanish, the majority are lexical in

nature. The number of phonological studies done on Cuban Spanish is

rather small, in relation to Cuba's linguistic importance in the

Caribbean area, and more generally, in American-Spanish dialectology.

Although there have been partial phonological studies done on different

areas of Cuba, no complete phonological study of any Cuban Spanish

dialect has ever been done. The following is a list of phonological or

phonetic studies done on some aspect of Cuban Spanish,

I. Published materials on Cuban Spanisht
1. Almendros 1958
2. Bartol 1965
3. Bartos 1966
4. Bartos 1970
5. Espinosa 1935
6. Haden and Matluck 1973 and 1974
7. Isbagescu 1968a
8. Isbasescu 1968b
9. L6pez Morales 1970
10. Terrell 1975a
11. Terrell 1975b
12. Terrell 1975c
13. Terrell 1975d
14. Terrell 1976a
15. Dihigo 1916
16. Montori 1916
17. Olmstead 1954

II. M.A. and Ph.D. theses on Cuban Spanishi
1. Bertot 1969
2. Clegg 1967

3. Guitart 1973
4. Lamb 1968
5. Sosa 1974
6. Vallejo-Claros 1970

III. Papers presented on Cuban Spanisht
1. Guitart 1974

IV. Unpublished studies on Cuban Spanisho
1. Terrell 1974a
2. Terrell 1974b

Because many of the above works are relatively unknown, they will

be briefly discussed in the remainder of this section, with emphasis on

the lesser known and more recent of these studies.

2.2.1 Published Materials on Cuban Spanish

In this section ten articles will be mentioned. Seven of these

will be briefly summarized, with an emphasis on a discussion of the

sections of each article that deals with the pronunciation of Cuban

Spanish. The remaining three articles, because they either were pub-

lished many years ago, or because of the paucity of phonetic material

they contain, will only be briefly mentioned in the last part of this


2,2.1.1 Almendros(1958)

This article is a thirty-seven page study, of which only ten pages

are devoted to the phonology of Cuban Spanish. The remainder of this

article is devoted to other topics, such as the historical origin of

Cuban Spanish, indigenous influences on Cuban Spanish,African influ-

ences on Cuban Spanish, etc. Lamb (1968) also contains a discussion of

Almendros (1958) in Chapter Two. Although Almendros collected his data

and wrote this article more than 20 years ago, when the linguistic

situation may have been very different from now, nevertheless, one must

be somewhat suspicious of some of his conclusions, since he apparently

relied to a certain extent on literary sources for some of his data, a

point also brought out in Lamb (1968).

Almendros (1958) devotes two pages to the discussion of vowels,

in which he mentions that vocalic segments in Cuban Spanish tend to be

more open than in Castilian Spanish, and then he mentions the follow-

ing phenomena 5

1. Change of /e/ to [Ci /*. -high
e.g., pasiar for jasear;t olpi6 for golpe6; tiatro for teatro;

siacab6 for se acab6.

2. Simplification of two consecutive identical vowels, e.g.,

alcol for alcohol. It is doubtful that his process does not occur in

all dialects of Spanish, at least in rapid speech.

3. Modification of certain stressed diphthongs, e.g., Daine for

]eine; sais for seis. Almendros (1958, p. 145) explains that "the

sound is not exactly reproduced as the vowel a, but rather more like a

sound between a and e."

4. Apocope of unstressed final vowels, e.g., mucho chico0 noch;

llueve; 2pcO. Almendros (1958, p. 145) explains that these vowels are

not completely apocopated, but that "some light traces of the vowel re-

main." He also notes that this process takes place especially "when

the preceding consonant is voiceless." Apparently what Almendros ob-

served here is a devoicing of this unstressed final vowel, which will be

discussed in section 3.8 of the present study.

Almendros (1958) devotes eight pages of discussion to the conson-

ants of Cuban Spanish. He discusses the following topics:

1. The relatively weak articulation of /ptkbdg/ in Cuban Spanish,
especially of /d/ and /g/, which in intervocalic position almost disap-

pear, but not completely.
2. The use of [h] instead of [x].

3. The archaic retention of Chi in some lexical items, e.g.,
Arz for hur; alar for halar.
4. The occurrence of seseo in Cuban Spanish.

5. The use of a pre-dorsal /s/ instead of the apical /s/ of
Castillian Spanish.

6. The phonetic realization of /s/ as Cs], [h], or [C.

7. The phonetic manifestation of /s/ / C being assimilated to
the following consonant, e.g., il-la for isla; avip-pa for avispa.
8. In words ending in -ci6n or -si6n the /s/ [']i, due to
the palatalizingg influence of the following Cy]," e.g., atenshi6n

Catens'y6Y for atenci6n [atensy6o]; profeshi6n [profe'sy6l for pro-
fesi6n Cprofesy6Si. I have never observed this phenomenon in MCS, nor

can I find any native speakers of MCS who will admit to having heard

Is] in this environment.
9. The assimilation of liquids to the consonant which follows
them, e.g., canne for care; aggo for algo; veddA for verdad; sewilleta

for servilleta; sukttame for sueltame. According to Almendros (1958),

this assimilation of liquids is "what most decisively distinguishes the

speech of Western Cuba from that of Eastern Cuba," taking place in the
three western-most provinces of Pinar del Rfo, La Habana, and Iatanzas.
10. In cases where /r/ is not assimilated to the following con-
sonant, its pronunciation in word-final position is not clear and

well-defined as in Spain, but ass "a.) a weakened sound;6 b.) with a

very brief aspiration with nasal resonance (mei. comei Cfor meior,

comer--R.M.H.1; c.) a sound intermediate between r and I por-l

conductor-l" (p. 150).

11. The occurrence of yefsmo in Cuban Spanish.

12. The phonetic manifestation of hq, 1 and v as a, e.g., aora
for ahora; gomitar for vomitar; cacagual for cacahual.

13. C[l -a [gI for the group hue, e.g., ievo [gw$o] for
huevo: g~ueco gwekol for hueco. Actually, what Almendros is showing

here is consonantalization, i.e., the glide [w] -> [ / # (see

Harris, 1969, p. 24).

14. The surface realization of palatal consonants as [f], e.g.,

Eato ama for chato; mar fr lamar.

15. The pronunciation of rr // much softer than in Spain.
16. "The x has a sound similar to that of the s" (p. 151).

17. A very brief mention of prothesis, epenthesis, aferesis,
syncope, apocope, metathesis, assimilation, dissimilation, and the pro-

nunciation of augmentative, diminutive, and deprecatory suffixes in

Cuban Spanish.

One serious weakness in Almendros' (1958) presentation of the

above phonetic data is that instead of giving phonetic transcription

for his data, he resorts to what is evidently some type of compromise

between spelling and phonetic transcription, thereby leaving the reader

to either make assumptions, or at times, simply guess at what he is

talking about. Bartos (1965)

In this six-page article, Bartos presents a three-page discussion

concerning various aspects of Cuban pronunciation and lexicology, with

emphasis on their causes, e.g., indigenous influences, African influ-

ences, Chinese influences, and other assorted influences. He briefly

mentions the fact that the intonation of Cuban Spanish differs from that
of the standard language, "perhaps because of the influence of the

Blacks in Cuba" (p. 144).

Bartoi (1965) presents a page and one-half of phonetic transcrip-

tion of two different passages, each shown in Castilian pronunciation

and Cuban Spanish pronunciation.

In the remainder of this article, Bartos discusses the following

phenomena found in Cuban Spanisht

1. The cantique found in the intonation of the Cuban Spanish of

the province of Oriente, i.e., the singing-like quality of this


2. The extraordinary rapidness of Cuban Speech, perhaps because

of the "frequent laxing" of unstressed vowels. Bartos suggests that

this rapidity of speech brings about phonetic changes or modifications

in the pronunciation of Cuban Spanish. It seems possible that the

opposite is true, i.e., the phonetic modifications give an impression

of rapidity to the uninitiated.

3. The occurrence of seseo in Cuban Spanish.

4. The occurrence of yelsmo in Cuban Spanish.

5. C#
/s/- 11J in Cuban Spanish.
6. /x/ -- [h] in Cuban Spanish.

7. The existence of the following ordered rules in Cuban Spanish:

a. V 7 / n

b. [ni -> Wf/7__.#

8. The simplification of consonant clusters in the pronunciation

of Cuban Spanish.
It can be seen from the above items that this article presents a

very limited amount of information on the phonology of Cuban Spanish, and

what it does present is rather well-known to anyone with a limited know-

ledge of American-Spanish dialectology. Bartos (1966)

This article devotes about one page to a discussion of the pro-

nunciation of the phoneme /b/ in Cuban Spanish. Bartog states that

during his stay in Cuba in 1961-62, he noted the occurrence of labio-

dental [v] in the speech of Cubans. On returning to Czechoslovakia, he

made a study of the speech of two Cuban informants, aged 25 and 26 years,

who were students there at that time. He observed the followings

1. X-ray photographs made by Barto" show that the stop realiza-

tion of /b/ is not limited to the environments of after a pause or

after a nasal, but that Eb] also occurs in intervocalic position in

Cuban Spanish, or in environments before [y], e.g., labio Clbyo].

2. X-ray photography made of the lexical items vava vaga and

6valo show the followings

a. When /b/ occurred in word-initial position, for the

grapheme v (vaval vaga), this segment was realized as bilabial, but at

the same time, the upper-front teeth lightly touched the lower lip, re--

sulting in a bilabial-labiodental coarticulation, e.g., vaa [t aya].

b. In intervocalic position, the phoneme /b/, when repre-

senting the grapheme v (6valo), was realized as a "purely labiodental"

(p. 98) articulation. In MCS, the realization of /b/ as labiodental

[V] occurred only in affected speech.

3. Bartos notes that, because of the limited sampling used in

this study, he cannot determine if the labiodental articulation [v] is

in complementary distribution with the other surface manifestations of

systematic /b/, or if [v] is merely in free-variation with these other

phonetic realizations. He does conclude, however, that the situations

concerning the articulation and distribution of the surface manifesta-

tions of systematic /b/ are different than as usually stated in

often-cited pronunciation manuals for standard Spanish. Bartos (1970)

This short article (three pages) brings up several interesting

aspects of the pronunciation of Cuban consonants. Some of Bartos' find-

ings in this article disagree with the findings of other authors who

have written about Cuban Spanish. As interesting as some of his find-

ings may be, Bartos (1970) has one notable weakness, in that all of his

conclusions are based on the speech of two educated subjects; "la pro-

nunciation de deux sujets cubains appartenant a la classes instruite"

(pp. 153-54)-

Bartos (1970) cites the need for the use of modern scientific

methods in the study of dialectology, and he discusses the influence of

the black and the indigenous substrata on the Spanish of Cuba. He also

suggests that the articulatory changes which he cites in this study may

lead to a "change in the articulatory system" of Cuban Spanish

(p. 154).

Bartos (1970) discusses the following phonetic data from Cuban


1. The occurrence of seseo in Cuban Spanish.

2. The phonetic manifestation of the sibilant [sI. Bartos cites

Canfield (1962, pp. 79-80) as stating that the [sI of Cuban Spanish is a

dorso-alveolar articulation. Based on X-ray photography "radiofilms" of
the speech of his two informants, Bartos concludes that the [s] of

Cuban Spanish, contrary to Canfield (1962) is articulated with the tip
of the tongue against the alveolar ridge or against the posterior part

of the upper incisors, "que la sifflante est prononcee avec la pointe

de la langue s*appuyant centre les alv6oles ou centre la parties

posterieure des incisives supgrieures" (p. 154). Thus he concludes

that the articulation of Cuban ls] is "apico-dental-alveolar." Bartos

also states that his X-ray films show the Cuban Cs] to be grooved,

rather than flat, "la forme du dos de la langue Stait lgerement con-

cave" (p. 154).

3. The occurrence of yvesmo in Cuban Spanish. His X-ray photo-

graphy shows no difference in pronunciation between the graphemes I
and y.
4. Systematic /x/ is realized phonetically as [h7 in Cuban

Spanish, rather than as [xl as in Castilian Spanish. Bartos states,

however, that the velar [x] is heard in Cuban Spanish "in proper names

of Germanic origin" (p. 154).

5. The phonetic realizations of /b/ in Cuban Spanish. Bartos
states, again based on his X-ray photography, that the phonetic comple-

mentary distribution of the allophones [b] and [$] does not occur in

Cuban Spanish as it does in European Spanish, i.e., [bl occurring after

a pause or after a nasal; [$ occurring elsewhere. Eartos (1970,

pp. 154-55) states that the "stop allophone [b] is not limited to the
environments of after a pause or after nasals, but that it is found

also in other positions, even in intervocalic position," but he adds
that "l'occlusion labiale n'est pas si forte e:t elle change selon les

voyelles voisines" (p. 155).

6. The existence of [vj in Cuban Spanish. Barto? states that his

X-ray photography proves the existence of the labiodental [v] as a pro-

nunciation for the grapheme v especially in word-initial and intervo-

calic environments. BartoY says further that this labio-dental pronun-

ciation is "rather widespread, and that it cannot be considered as a

pronunciation of school masters."

7. The phonetic realizations of [i0. BartoY finds great articu-

latory variation in the pronunciation of this segment, and he states

that "the dorsum of the tongue is clearly concave" in the articulation

of this consonant, although Navarro-Tomgs has stated that the back of

the tongue is convex.

8. The stop consonants [ti and Cd], usually dental in standard

Spanish, are more nearly alveolar articulations in Cuban Spanish.

9. The occurrence of [C7 in Cuban Spanish, which, at the end of

words, nazalizes the preceding vowel, "with a tendency for the nasal

consonant to disappear."

10. Bartot notes that Cuban pronunciation is characterized by a

nasal quality.

The type of study presented here by Barton, using X-ray photo-

graphy, is certainly to be commended. Much more research of this type

needs to be carried out. It is unfortunate, however, that Barto' used

only two informants, a fact that somewhat diminishes the credibility of

his findings. Bartot also fails to state whether his two informants

resided in Cuba or elsewhere.

2,2.1.5 Espinosa (1935)

In this article, Espinosa discusses the phonetic manifestations

of several consonantal graphemes of Cuban Spanish. In discussing this

article, only those data of Espinosa that concern Cuban pronunciation
will be discussed. One serious weakness of this study, also pointed

out in Lamb (1968, pp. 13-14), is that Espinosa's observations do not

appear to be based on any specific group of informants, but rather on

the author's personal observations on Cuban Spanish, without regard to

geographic area, socio-economic level, or educational level, as though

Cuban Spanish were a highly homogeneous entity,

Espinosa provides the following observations about the pronuncia-

tion of Cuban Spanish:

1. The allophone [s] of /s/ in intervocalic environments.
Espinosa describes [s] as a dental, rather than alveolar, sibilant. He

also refers to [sJ as "sonoro" (pp. 8-9). Although this word normally

means 'voiced' in Spanish phonetics, one assumes he is using this term

to refer to an impressionistic observation, in the more traditional use

of "sonoro," meaning 'clear.'

Espinosa states that within a word /s/ [h] / C. He

states that this phonetic manifestation of s is similar to that of the

grapheme j, "without arriving at being such a guttural sound" (p. 8),

Espinosa says that the phoneme /s/ at the end of a word is very
weakened, with a marked tendency toward deletion.

2. /r/ -9 [h] / __ b g k 1. m n pl, thereby leveling the
phonemic distinction between /r/ and /s/ before these consonants, e.g.,

casco and cargo would both be pronounced [kthol.7

3. Both /s/ and /r/ in preconsonantal position within a word may
even become deleted.

4. [q] has been lost in Cuban Spanish, although this pronuncia-
tion persists in the speech of Spaniards living in Cuba. In Cuban

Spanish this [e] appears as [s h ]j. It is difficult to follow

Espinosa's exposition, especially in this section, but he refers to the

"phoneme" /z/ of peninsular Spanish, which he is apparently using as a
symbol for [e]: "el fonema 'z' se ha extinguldo completamente en el

habla criolla, a pesar de su persistencia sonora en la fongtica de los

espaeoles, que en gran nimero resident en el pals" (p. 11).

5. /d/ has several different phonetic manifestations and has an
effect on neighboring vowels.

/d/ -. [C in word-final position, at least in everyday unaf-

fected speech, in words stressed on this last syllable. Espinosa

states that after the deletion of this word-final /d/, the preceding

vowel becomes lengthened. If, however, this word-final /d/ is not de-
leted, part of the length of this preceding vowel is absorbed by the

word-final /d/. Espinosa also states that "when the word is stressed

on the penultimate syllable, the final 'e' is scarcely pronounced,

after the 'd' is lost in pronunciation,"e.g., cesped (p. 13).

Espinosa states that there is a tendency for /d/ -+ [C /

V V. After deletion of this intervocalic /d/, the preceding vowel

becomes strengthened, and the following vowel weakened if these vowels

are not identical. "If these vowels are the same, e.g., nada, they

fuse into one vowel, the word loses a syllable and the other sound be-

comes lengthened, i.e., [ni]" (p. 13).

6. "The phoneme of the letter j" (p. 14) has completely disap-

peared in word-final position in Cuban Spanish, but remains in word-

medial position. Espinosa makes no mention of this "phoneme" in

word-initial position, nor whether its allophonic manifestation is [x]

c~ [h]. In this article, Espinosa makes repeated references (p. 14 and

ad passim) to a tendency of Spanish to delete, or rid itself of, the

noun case-inflections of Latin, accounting for the tendency of Spanish

in general, and in Cuban Spanish in particular, to delete word-final


7. The labio-dental [v] does not exist in Cuban Spanish, except
in affected-speech.

8. The graphemes 1 and v are phonologically leveled to /b/,
with the stop allophone [b] occurring after the bi-labial nasal /m/, and

the continuant [C6 allophone generally occurring elsewhere.

9. The occurrence of yetsmo in Cuban Spanish.

10. The phoneme /1/ has the following allophonic manifestations

a. /1/ 1 ___

b. /1/ -3 [1 / V $, and this preceding vowel undergoes
compensatory lengthening.

c. /1/ [] / -stressi __ e.g., marmol [mrmo].
d. the confusion, of /1/ with /r/ in word-medial and
word-final positions, and, less frequently, the confusion of the /r/

with /1/ in these same positions.

Following his discussion of the ten different items of pronuncia-

tion listed above, Espinosa mentions that the phonological environments

of these different consonantal phonemes may also cause other allophonic

manifestations. He then states that all of these phonological changes

from standard Spanish to Cuban Spanish, e.g., /d/ -, [f] / #, show

a tendency toward articulatory simplification, what is known as 'the

principle of least effort.' Espinosa should have stopped at this

point in his exposition. Although his observations are based, evi-

dently on his own general impressions of Cuban pronunciation, and not on

a scientifically controlled body of data, his phonetic observations, and

his observations about this articulatory principle of least effort are,

for his time, quite commendable. He proceeds, unfortunately, to the

following ad hoc observations as to the cause of this principle of

least effort; "In that principle which is manifested in languages, but

adquires alarming proportions in the Spanish spoken in Cuba, constantly

reinforced by the great tendency toward laziness, which is innate in

Cubans, perhaps because of determinate influences of climate" (p. 18).

Espinosa goes on to observe that these phonological changes (which

proceed in the direction of least articulatory effort) do not affect

word-final vowels, "which are always pronounced, although at times so

lightly, that the ear of foreigners, although they know Spanish, does

not perceive nor distinguish the clear [sonoras] blendings which dif-

ferentiate a from o, nor e from i at the end of words" (p. 19).

Apparently what Espinosa is saying is that although non-Cubans observe

changes in the phonetic manifestations of word-final vowels, Cubans do

not perceive these changes.

These types of statements on the part of Espinosa cast a severe

shadow of doubt on the credibility of all his observations, since he is

obviously capable of basing his judgement on what must be considered

less than a pure scientific approach to the analysis of his data.

Aside from the ad hoc conclusions that Espinosa finds it necessary

to draw after his presentation of his observations on Cuban Spanish,

this article is interesting, because it shows that many traits of Cuban

pronunciation, supposedly only very recent, have indeed been present for

over forty years. Espinosa's article is written in such a way that it

is quite difficult to read in some parts, and in other parts it is

impossible to decipher the author's thoughts. One thing that makes this

article difficult to follow is that Espinosa either does not give

examples of what he is discussing, or when he gives them, he does so

with the spelling system of Spanish, completely avoiding the use of

phonetic transcription. Haden and Matluck (1973 and 1974)

These two articles will be discussed together since they deal with

the same body of data, with Haden and Matluck (1974) being basically an

English version of their 1973 Spanish article.

The data for these studies wre gathered in Miami, Florida using

recently arrived Cuban refugees from La Habana. The corpus of data col-

lected consisted of 400 hours of speech, from 454 interviews involving

750 informants. According to the authors, the informants were all "pro-

fessionals of one kind or another" (1974, p. 20). From these 400 hours

of taped material, 100 hours were selected for analysis. The authors

report that "Phonetic transcriptions were made from three-minute sec-

tions preselected from each of sixty of these" (1974, p. 20).

It should be pointed out here that the data used in this study

were collected as part of a larger study, "a global research project

sponsored by the Inter-American Program for Linguistics and Language

Teaching and entitled A Coordinated Study of the Linguistic Norms of

the Principal Cities of Ibero-America and the Iberian Peninsula (1974,

p. 20), and not specifically for Haden and Matluck (1973) or (1974).

This same corpus of data has been used for many other recent studies

done on Cuban Spanish, such as Vallejo-Claros (1970) and a number of

Terrell's recent studies on Cuban Spanish.

The greater portion of Haden and Matluck (1973 and 1974) gives a

detailed analysis of the suprasegmental phonology of Havana Spanish.

The following observations of interest to the present study, are offered

by Haden and Matluck relative to the segmental phonology of Havana


1. As is typical of Caribbean Spanish, the vowels of Havana

Spanish are relatively stable, while the consonants show a much higher

degree of variability.

2. Vowels in Havana Spanish weaken only in absolute-final en-

vironments. This weakening occurs most frequently in rapid, familiar

speech. The authors report that this vocalic weakening is limited to

/e/ and /o/, and occurs as a devoicing of these vowels, e.g., campesino

[kampesfnol, mexicano [mehikanol. This phenomenon has, of course, been

reported previously in most phonological studies on Cuban Spanish, but

as usually occurring after palatal consonants, and not after /n/.

3. Haden and Matluck provide the well-worn hypothesis that the

vowels in Havana Spanish "tend--very generally speaking, of course-to

open somewhat in blocked syllables and to close slightly in free

syllables, without regard, in either case, to the nature of the conson-

ant that follows. .. Cthel phoneme regularly opens appreciably before

a following /s/ which has been either aspirated or dropped, resulting

thereby in a significant contrast between second and third-person verb

forms" (1974, p. 21), e.g. v [be"l vs. yes [bel. As has often been

the case in the past with this often-repeated hypothesis, the authors

offer no empirical evidence to support this theory. As has been re-

ported in Hammond (1973), no evidence, either acoustic or perceptual,

could be found in MCS to support such a theory.

4. Non-identical contiguous vowels are rigidly maintained, i.e.,

not weakened or reduced, except that there is a "slight tendency for the

/a/ to weaken in tension or be eliminated when followed by /e o/,
especially so in proclitic position . ." (1974, p. 21). The authors

offer the following examples of this weakening of /a/s la enterrlbamos

5. The authors report strong vowel nasalizatin in the Spanish of
La Rabana
Sa. Before the velar nasal in absolute-final position, e.g.,
avion [apyozj!{ lien [byenj|
b. In syllable-final position before a nasal consonant fol-,_
lowed by /s/, e.g., entonces [entese].
6. The Spanish of La Habana, like other Caribbean dialects is
"Notorious among Hispanic phonological systems for the weakness in the
muscular tension with which consonants are articulated, not only in
Final position, but in intervocalic position as well" (1974, p. 21).

7. Word-final /d/ is deleted with remarkable consistency, and
intervocalic /bdg/ are often deleted in Havana Spanish.
8. Almost 100% lack of the voiced labio-dental fricative as an

allophone of /b/, contrary to what was reported in Bartos' (1966) and
Isbatescu (1968b).

9. /s/ in intervocalic position is mentioned as a voiceless
sibilant. In extremely rapid speech, however, "and in close proximity
to other sibilants, it will at times be partially voiced (1974,
p. 22), e.g., ese sofrito [esesofrfto], where Cs] represents a par-
tially voiced alveolar sibilant.
10. In syllable-final position, followed by a consonant, /s/ ap-
pears phonetically ass [h- if the following consonant is voiceless,

e.g., espera [ehperal; [fin if the following consonant is voiced, e.g.,

mismo EmFimo]; or C1 in either case. Haden and Matluck generalize
That the faster the tempo of speech, the more frequently [Cl is heard"
(1974, p. 22).
11. In word-final environments, followed by a vowel, /s/ appears
as either [hI, [91, or C[.
12. In absolute-final position [sj, [hl, [C7] and [Cf are in free
variation as phonetic variants of /s/.
13. The authors report [fl as the only allophone of the phoneme
/f/, and found no evidence of the voiceless bilabial spirant [fC, which,
as reported in Chapter Three of the present study, does occur in MCS.
14. The phoneme /r/ between vowels and in consonant clusters of
obstruent plus /r/, appears regularly as [r]. /r/ will rarely and
sporadically appear as [j] or [Cj in these environments in very rapid
speech and in certain lexical items, e.g., para all [Cpayl.
15. In syllable-final and word-final environments, /r/ has the
following allophoness [r], [r], [rJ, and [Cs] (a lengthened assibilated
liquid). In syllable-final position, the allophone [Cr appears most
frequently. Haden and Matluck also report some occurrences of [CJ,
(which they describe as a voiced lateralized fricative), in syllable-final
and phrase-final environments.
16. In absolute-final environments, the authors report the [C
allophone of /r/ predominates, sometimes devoicing and sometimes acquir-
ing a nasal resonance.
17. The phoneme /V/ appears normally as [71 between vowels, some-
times devoicing. In initial position /r/ may appear as the flap Cr[
or may become devoiced, [fC. As will be shown later in Chapter Three of
the present study, the [rl and [r] allophones of /P/ also occur in KCS
rapid speech.

18. The nasal /n/ is always velar in phrase-final position, be-

fore any pause, and word-final followed by any vowel; Haden and Matluck,

however, report that a nasal regularly assimilates to the place of ar-

ticulation of a following consonant, as in standard dialects. Also,

the authors report that /n/ is lost before /s/, "leaving traces of its

nasality in the nasalization of the preceding vowel" (1974, p. 24),

e.g., entonces [ent'se J. Isbagescu (1968a)

This publication gives a rather long detailed account of the

Spanish spoken in Cuba. Isbagescu's data is based on information ob-

tained from six informants, who were students in BucareSt in 1964, when

this study was done. These six informants ranged in age from twenty to

thirty-two years of age at the time of the study. Three of these in-

formants were from the Cuban province of La Habanal one was white, one

black, and one mulatto. The remaining three informants were whites from

the provinces of Matanzas, Camaguey, and Oriente. Isbasescu had no

informants from the Cuban provinces of Las Villas or Pinar del Rfo,

The author provides no information as to how long these informants had

lived outside of Cuba.

Lamb (1968) includes a summary of the data found in Isbasescu

(1965), which provides the corpus of data found in Isbagescu (1968a).

One of the negative aspects of Isbagescu's data, also pointed out by

Lamb, is that it makes sweeping generalizations about Cuban Spanish,

based on a small sampling of data from informants representing four of

the six provinces of Cuba.

Isbasescu (1968a) contains approximately thirty-five pages of

discussion of the pronunciation of Cuban Spanish, with ten pages devoted

to vocalic segments and twenty-five pages to consonants. Prior to this

discussion, she provides background information concerning Cuban Spanish.

The last half of this publication provides a sixty-seven page phonetic

transcription of the author's interview with her six informants.

Isbasescu (1968a) will be discussed briefly below, limiting the discus-

sion to her data on Cuban pronunciation, and largely to those points

which are not already well-known about Cuban Spanish, or that are in

disagreement with information provided by other sources.

In her discussion of the vocalic segments of Cuban Spanish,

Isbasescu provides the following information:

1. Contrary to other sources which report that the vowels of

Cuban Spanish tend to be more open than those of standard Spanish,

Isbasescu reports that her data show that this "greater open quality of

vowels is closely related to the position of the vowel within the word;

i.e., that it occurs more frequently in word-medial vowels, especially

in stressed ones, while unstressed vowels and vowels situated in

word-final position, on the contrary, are characterized by an obvious

closed quality" (p. 22).

2. The vocalic phoneme /a/ has two surface variants, [a] a pala-

tal phone, and [a] a velar allophone. Isbasescu found that in Cuban

Spanish, there was no tendency toward phonemicization of these allo-


3. /a/, as well as other vocalic phonemes, has a weakened

variant, "which appears in weak position, conditioned by the accent of

the word" (p. 22). This weakened variant occurs in word-initial posi-

tion, hasta tal; within a word, especially when /a/ is in post-tonic

position, platano [plAtanol, and most frequently in word-final

position, tasta Cpstal. Isba-escu also reports that at times this

weakening of word-final /a/ causes a weakening in the entire final
syllable, e.g., guanabana [gwanfana", or this word-final /a/ may com-

pletely disappear, e.g., siembra [syembrj. All of Isbatescu's data on

vowel weakening is, of course, contradictory to the information provided

on vowels in Cuban Spanish by Espinosa (1935).

4. The vocalic phonemes /e o i/ each have two surface variants,
e o i] more open, and [e o i], more closed. Isbasescu reports no

phnemicization of these open and closed surface variants. (Hammond,

1973, showed that not only was there no difference of vowel quality, but
that in syllable-final position within a word, vowel quantity and not
quality had become phonemicized.) She also reports weakened variants of

/e/ and /i/ in unstressed positions, and a weakened variant of /o/ only

in word-final position.

5. The vocalic phoneme /u/, as reported by Isbagescu, has only
the allophone [u], and does not have the open and closed allophones that
occur in the three [-lowl vowels. /u/ is also reported to have a

weakened variant, as was the case with all other vocalic phonemes, oc-

curring in unstressed positions.

In her discussion of the pronunciation of the consonantal phonemes
of Cuban Spanish, Isbasescu (1968a) presents the following data

1. The labio-dental [v], as an allophone for the phoneme /b/, was
heard on occasion by Isbasescu, but she reports that her investigations
into this phenomenon "are still insufficient to allow us a deeper analy-
sis of the problem" (p. 34). This problem is discussed in Isba escu


2. The stop and fricative allophones [bJ and [C of the phoneme

/b/ are not in complementary distribution in Cuban Spanish, but rather
they are in free-variation.

3. /b/ may appear phonetically as [wl, la boca [law6ka], Cuba
CkGwa6 asaba [passwa], i.e., when /b/ appears within a word, or within
syntactic mnits such as NP's; /b/ may appear as [w / $_ [Cr, when
the preceding syllable ends in /o/, e.g., sobre [s irel, obre
Cp6wrel; /b/ may also be deleted in intervocalic environments.
4. As was the case with /b/, IsbaSescu reports that the allo-
phones [d] and [f of /d/ are not in complementary distribution in Cuban
Spanish, but they are rather in free-variation. The author also reports
the frequent loss of intervocalic /d/.
5. /k/ -. [C] / V__ e.g., la casa [laiasa], Eacardf
C$agarfil. Isba escu claims this process of lenition to occur only with
/k/, but not with the other two voiceless stops /p t/, as reported in
Saciuk (1974).
6. /k/ -3 [C when followed by a diphthong, e.g., cuando
7. As was the case with /b/ and /d/ the stop and fricative
variants of /g/ do not occur in complementary distribution but rather in
8. /g/ [9n only rarely in intervocalic position, contrary to
what has been reported by others.
9. /f/ may appear as [ intervocalically in word-initial posi-
tion within a syntactic unit (such as an NP), e.g., otra forma [otra
10. /h/ occurs in Cuban Spanish instead of /x/ of the standard
language. The phoneme /h/ has a weakened variant, sometimes completely
disappearing in intervocalic position, e.g., coge Ck6heT, muier [muer].

Isbajescu reports that /h/ may appear as [C] in intervocalic position,
e.g., frijoles [frilSles].
11. In addition to the allophones Cs h $1 of the phoneme /s/
usually reported for Cuban Spanish, the author states that /s/
[( / E [dl, e.g., pue ramos decir Cpudyramo gesfr]. Isetaescu
reports that when /s/ [~ within a word, this deletion may be "at
times accompanied by a lengthening of the preceding vowel," e.g., esto
[Dto] (p. 46), as was reported in Hammond (1973). Isbasescu states,
however, that this compensatory lengthening "is not produced in most
cases" (p. 46).
12. [CE and [71 appear in free-variation as allophones of /c/,
this variation being more likely to occur-word-medially rather than
13. The allophones Cyl and [C of the phoneme /y/ appear in free
variation in Cuban Spanish, with Cy] occurring more frequently.
14. The dorso-velar nasal consonant [C] occurs very frequently in
Cuban Spanish, and can be found in word-final position, followed by a
word whose initial segment is a consonant, e.g., en pellcula
[Npelfkula], con iam, n [koham!,i; also [3! occurs in absolute final
position, e.g. en y, g [by limon Climoq]; [C11 is heard in word-final
position when the next word begins with a vocalic segment, e.g., en agua
Cagwal], son altos [si8 ltos7.
15. /n/ may appear as a very weak articulation intervocalically,
e.g., animal animal1], cuando nace [kwC1ndonse], and /n/ may occa-
sionally be deleted, e.g., de ninguna [deyng6nal; word-final /n/ may
also nasalize a preceding vowel and then be deleted, as shown in the
following rules,

(a) V -) 7/___ /n/
(b) /n/ -0 [C _-
Rule (b) occurs after rule (a), in a natural feeding-order, with rule
(b) applying only to the output of rule (a), e.g., con 1 [iK5el, son
unas [snasl; Isbasescu also reports that /n/ may infrequently appear
as C1], e.g., ante [9ltel, lim6n [lim6ll (p. 51).
16. /r/ and /r/ do not occur in complementary distribution
everywhere except in intervocalic position in Cuban Spanish, but rather
in free-variation, including word-initial position, e.g., radio [59yo],
but recoge [rek6he].
17. /r/ in Cuban Spanish is "generally much softer (mas suave)
. . than in the Spanish of other regions, so that it almost never has
more than two vibrations" (p. 52). The /r/ of standard Spanish has a
minimum of three vibrations.
18. The assibilated [Cr appears in intervocalic position as a
surface form of either /r/ or /i/. Isbagescu also reports [r1 in
word-initial position, e.g., refinada [refingfal [refinaia] (p. 53).
19. /r/ -- [ /L__.~ e.g., az.car [aska;] /r/ -* [$1 /
$, porQue [p6kel; /r/ may also be deleted when it occurs within a
syllable, e.g., trae [tie].
20. /r/ in syllable-final position may be assimilated to a fol-
lowing /n/, e.g., came [kane]; /r/ may be less frequently aspirated,
e.g., transporte [thnhp6rtej; even less frequently, /r/ may become
vocalized to [i], e.g., hacer grano [aseigrfno].
21. /?/ -9 [kr] /#, e.g., roaa [kr6pa], refresco
[krefrghko] (p. 54).
22. The velar allophone /R/ was not observed in Isbagescu's data.

23. /1/ may appear as [j both in syllable-final position and
within a syllable, e.g., cualauier [kwakykrl, planchar [paEfnr1.
24. The neutralization of /r/ and /1/ occurs in all areas of
Cuba. This neutralization takes place in word-final position,
syllable-final position, within a syllable, and intervocalically. Both
/r/ and /1/ ay appear as [il, Cr], or [li. Isbasescu also reports that
both /r/ and /1/ are pronounced successively in word-final position,
e.g., ser [serli, decir [desfrl].
25. The archiphoneme resulting from the phonological neutraliza-
tion of // and /y/ (yeismo) has three allophones, occurring in
free-variation [~], a palatal lateral; [yl, a palatal fricative; [y],
a palatal affricate, e.g., llamando .1amando]; llena [ygna]; gall

[gafol. All of Isbagescu's examples contain -I in their spelling, and
she does not discuss how words with 2 in their spelling, e.g., ava,
would be pronounced (p. 59). After approximately six years of constant
contact with ICS, I can inequivocally state that I do not ever recall
hearing a palatal lateral from any speaker of this dialect. Isbasescu (1968b)
This five-page article devotes three pages of discussion to the
historical development of the sounds [b] and [v] in Spanish, and how the
graphemes b and v are pronounced in different dialect areas of the
Spanish-speaking world.
Isbasescu reports that E[v appeared in the speech of all six of
her informants (this article uses the same corpus of data as Isbasescu,
1968a). Their labio-dental pronunciation generally corresponded to the
grapheme v, but was also found corresponding to the grapheme b, e.g.,
resbala [rehvala], se baila [sevayla]. Isbasescu states that C[b, [E1,

and [v] occur in free-variation in Cuban Spanish, son viviendas

lyivyinda I, vamos [9mos], uva (Gba], rbol [grbol], universidad
[universida], servir serverr], bueno [vweno], voy [vbyi.
Isbasescu concludes that the graphemes b and v in Cuban Spanish
may be realized phonetically as [b], C, or [v], these three sounds be-

ing in absolute free-variation. L6pez Morales (1970)
In this book, L6pez Morales devotes approximately twenty-eight

pages to the pronunciation of Spanish in Cuba. Some of the data con-
cerning Cuban pronunciation is the following:
1. In emphatic everyday speech, when /d/ has been deleted, the
preceding vowel undergoes compensatory lengthening, e.g., cansado
[hansatol (p. 109).
2. In careful pronunciation, /s/ is realized as a convex
dorso-dental-alveolar segment, usually voiceless. Aspiration of /s/
occurs more frequently in Western Cuba, before any other segment. This
aspiration of /s/ may be voiced before /b d g 1 n/. L6pez Morales re-
ports that /s/ --> [0] within a word is infrequent, "although at times
the aspiration is difficult to perceive" (p. 110).

3. When /s/ -9 [CC in word-final position, the more open quality
of the preceding vowel is observed. As previously mentioned, it was
shown in Hammond (1973) that in MCS there was no change involved in
these vowels in word-final environments.
4. /1/ in word-final and syllable-final position is devoiced, but
not deleted, e.g., carcel [karsel1i sol [s61] (p. 111).

5. /r/ and /1/, in everyday speech in the cities, may be pro-
nounced as a lax, voiceless apico-alveolar fricative. Likewise, /1/

may be produced as a voiceless nasal aspirate, e.g., Tapel [papeh];
sal sahj.
In everyday urban speech, /r/ is realized as a voiceless nasal
aspirate, e.g., venir [benfh]; comer [komeh].
6. /r/ in final position is usually produced as a fricative,
either voiced or voiceless, with minimal tenseness.

7. The normal surface realization of /k/ in the group /kt/ is

[,E9 e.g., doctor [doit6r]; cartcter [karIter]. "The realization [of
/k/ in the cluster /kt/] as a voiceless stop is highly affected"

(p. 111).
8. In everyday speech the cluster /rd/ is realized phonetically
as [dd], with the /r/ devoicing and assimilating to the following /d/,
e.g., verde [bedde]. This according to Lopez Morales is the only fre-
quent obstruent assimilation to a following consonant that occurs in
Cuban Spanish.

9. [1,9 a bilabial voiceless fricative, occurs in Cuban Spanish
as a free variant of the phoneme /f/, e.g., fiesta [Cyesta]; facil

[Ps"I'; fo UpsT,
10. When /s/ is realized as [s] its most common articulation is
predorsal dental-alveolar, voiceless, with the body of the tongue con-
11. /x/ is never realized as [xJ.
2.2,1.10 Terrell (1975a)
In this study, Terrell uses some of the same data which will be
presented in Terrell (1974b) to support his proposal for a theory of
how the study of the phonologies of different dialects should be ap-
proached. Since these data from Terrell (1974b) are discussed in
section 2.2.4 of the present study, they will not be repeated here. Terrell (1975b)

In this study, which uses Cuban Spanish as a data source, Terrell

argues that there are functional constraints on word-final s deletion

which are correlated with redundancy in surface structure. The data

used in this study are some of the data which will be presented in

Terrell (1974b), and since these are discussed in section 2.2.4 of the

present study, they will not be repeated. Terrell (1975c)

In this study, Terrell obtains his data from the tapes made for

the "Proyecto del studio coordinado de la norma lingiistica culta de

las principles ciudades de Hispanoamerica y de la Peninsula Iberica."

The informants used in these interviews are all natives of the Cuban

province of La Habana. Terrell (1975c), primarily a theoretical study,

provides the data given below relative to nasals in Cuban Spanish.

In syllable-final environments within a word, nasal assimilation

occurred 84, of the time, while nasal deletion with prior nasalization

of the preceding vowel occurred 16' of the time, i.e., less than 15 of

the time a velar nasal occurred in this environment.

In Chapter Three of the present study (Tables VIII and XI), nasal

assimilation is reported 31.0 of the time, the velar nasal 42.2% of

the time, and /VN/ 4 [VI 26.87. of the time in syllable-final


The percentages reported by Terrell(1975c) and those reported in

the present study for surface manifestations of nasals in syllable-final

environments within a word are extremely different.

Likewise, if we compare the data given in these two studies for

nasals in word-final environments within a breath-group, we again see

radical differences in the percentages reported.

Terrell (1975c) reports the alveolar nasal occurring 3 of the

time, nasal assimilation at a rate of 335, the velar nasal 2,6 of the

time, and /VN/ 4 [7V1 occurring 38" of the time-all in word-final
environments within a breath-group.

In Chapter Three of the present study (Tables V and VII), the
following percentages are reported in this same environment: nasal
assimilation 3.L4, the velar nasal 62.59, /VN/ : [V] 33.7%.

There are a number of variables that could account for the dif-
ferences in the percentages given in these two studies, e.g., interview

environment, age, type of interviews, etc. Also, as discussed in sec-
tion 3.3 of the present study, it is often extremely difficult to dis-
tinguish nasalized vowels from velar nasals.

It should be pointed out, however, that the percentages for

phonetic occurrence of the velar nasal reported in Terrell (1975c)
(26^ in the environment / __. and Q0 in the environment / $),
especially his syllable-final percentage, differ radically, with the
exception of Haden and Matluck (1973 and 1974), with every other study

done on Cuban Spanish, e.g., Guitart (1973), Isbagescu (1968a), etc.
Also, the citing in Terrell (1975c) of less than 1 occurrence of the
,clar nasal in syllable-final environments is intuitively unacceptable
with my own personal experience with Cuban Spanish and MCS.
Regarding deletion of the nasal in syllable-final environments,
Terrell (1975c) offers the following data, nasal deletion occurred 71Z
of the time before voiceless fricatives /f s x/, 10l, of the time before
the voiced fricatives [ff i C, and 2V of the time before the voiceless

stops and affricates /p t k 6/. Terrell makes the point here that nasal
deletion "follows a scale of consonantal occlusion" (1975c, p. 8).

In word-final environments before consonants, Terrell (1975c)

reports nasal assimilation or alveolar [n] occurring 6Q0 of the time,

the velar nasal 3% of the time, and nasal consonant deletion 395 of the

time. In word-final environments before vowels, Terrell's data show

nasal assimilation or alveolar [n] as occurring at a rate of 3r, the

velar nasal 59/, and nasal consonant deletion 38%. In word-final

position, Terrell reports that nasal assimilation of the alveolar [n]

occurs at a rate of 8W, the velar nasal at a rate of 54V, and nasal

consonant deletion at a rate of 38&o.

Terrell observes that "The most striking thing about these data

is the fact that in word-final position, velarization occurs before

vowels and before pauses with the same frequency as nasal assimilation

before consonants" (1975c, p. 8). Terrell (1975d)

In this study, Terrell uses data from Cuban Spanish in an effort

to argue for the superiority of natural generative phonology over what

he calls abstract generative phonology, i.e., the standard generative

model. This study, however, is largely theoretical in nature. What

data that are presented from Cuban Spanish will be given in Terrell

(1974b). These data are discussed in section 2.2.4 of the present


2,2.1.14 Terrell (1976a)

This study summarizes some of the findings in Terrell's earlier

studies relative to different factors affecting s deletion in

word-final environments in Cuban and Puerto Rican Spanish.

The data from Cuban Spanish given in this article are presented in
Terrell (1974a) and (1974b). Since these data are discussed in section
2.2.4 of the present study, they will not be repeated here. Other Published Materials
on Cuban Spanish
In this section, three articles will be very briefly discussed,
two of them because they were published many years ago, Dihigo (1916)
and Montori (1916), and contain very little information about the pro-
nntciation of Cuban Spanish, and the third, Olmstead (1954), because
it only contains one item relevant to the present study. All three of
these articles are also discussed in Lamb (1968).
Dihigo (1916) reports the following data on the pronunciation of
Cuban Spanishs

1. /b/ -4 [gJ /_ -low e.g., abuela [aKwglaT;
Lback, volver [golver]

2. /s/ [n], e.g., resfriado [renfryfo!l; [r], e.g., las [larl;
[i], las [1ai]; [,> historic [it6rya!.
3. /k/ [rJ], e.g., contract Ioontrartol.
4. /d/ -. [1, [rn or [ C, e.g., admirar [almirr]r; periodico
[pery6rico]; miedo C[yeo!.
5. /f/ -4 [xj, fecha [xeca,.
6. /g/ -> [x, e.g., agu.iero [axuxero].
7. // Ci], e.g., vuelvo Cbweio].
8. /n/ -4 [rI or C11, e.g., conmigo Ckormnno] and [kolmfgo].
9. /r/ [Cn, [s i or [i], ege, viraen Cbinxen]; came
C[ksnell cuerpo Ckweipo].
10. The following instances of vowel raising /a/ -. Ce], e.g.,
astilla [estlyal; /a/ C [u], e.g., trancada [trunkgalf /e/ -4 Ci],

e.g., Peor [pyorlj /e/ [ul, e.g., terrn turnin; /o/ -) [u],

e.g., cn [Ckuml.
11. The following instance of vowel lowering, /i/ C [e],

e.g., mismo [mCmo].
12. The following example of vowel fronting: /o/ -C [e],

e.g., oSOLOS esemos].
Dihigo's study, as was the case with Espinosa (1935), was appar-

ently :based on the author's general observations about Cuban Spanish,
and not on any scientifically controlled body of data. It is unclear

as to whether his data is supposed to be representative of normal Cuban

speech, low class speech, etc. The phonetic transcriptions provided

above are my interpretation of Dihigo's graphemic representations of my

understanding of his data, as his work was without phonetic transcrip-

tion. It should be mentioned, however, that if the processes described

above (numbers 1-12) from Dihigo were phonological processes of Cuban

Spanish in 1916, they are not phonological processes of KCS now. While

it must be admitted that items such as (1), parts of (10), (11), and

(12) may rarely be heard in MCS today, other processes such as (2)

through (9) can only be described as ludicrous for MCS speech, as

spoken by any normal member of this dialect area.

Montori (1916) contains only little more than a page of text devot-

ea to Cuban pronunciation, and Montori took all the data he dis-

cusses directly from Dihigo (1916). Therefore, no further discussion

of Montori (1916) will be given here.

Olmstead (1954) is a two-page article about several aspects of

pronunciation found in the Spanish spoken in Regla, Cuba. He presents

one observation relevant to the present study: /s/ -4 [s h 1/ C.

When /s/ -> [j / C, there is a compensatory lengthening of the

vowel preceding this deleted segment.

2.2.2 M.A. and Ph.D, Theses on Cuban Spanish

In this section two M.A. theses and four Ph.D. dissertations done

on Cuban Spanish will be discussed. The greater part of the discussion

will be devoted to Lamb (1968), because this study contains by far the

largest and most detailed discussion concerning the pronunciation of

Cuban Spanish. Only those phonological or phonetic observations not

discussed in Lamb (1968) will be mentioned in discussing the remaining

five works. Of these five, Guitart (1973) contains the greatest amount

of data not discussed in Lamb's work. It should be pointed out, how-

ever, that in most of these studies, the primary purpose of the author

was not that of presenting data, but rather, discussing theoretical

implications, lexical data, or sociological occurrences of the data

presented. Bertot (1969)

This thesis is primarily a listing of the allophonic manifesta-

tions, along with their environments, of the vocalic and consonantal

phonemes of Cuban Spanish. Bertot also discusses how these allophones

differ in these environments from the allophones of standard Spanish in

the same environments. Bertot (1969) also includes some discussion of

the historical background of Cuban Spanish, and a brief discussion of

dialectical differences among her informants, and how the allophonic

distributions of her data correlate with the findings on Cuban Spanish

later included in Resnick (1975). Clegg (1967)

This short M.A. thesis contains one and one-half pages of text and

five pages of tables. In this study Clegg used as data the spontaneous

speech of one 59-year-old female informant from Havana, Cuba. This in-

formant had completed secondary education and had lived in the United

States for four years. From a corpus of more than one hundred items,

Clegg chose forty-five for the spectrographic analysis. The author

analyzed and compared the first and second formant frequencies of the

vowels /a o e/ in stressed position, unstressed position, and in

unstressed position before /s/ -9 [h]. In all cases, this /s/

[lh was the plural morpheme on a noun or adjective.
The only significant difference reported by the author was that

the average frequency of the first format for the vowel /a/ followed by

/s/ -- [h1 was 156 cps lower than the first formant of /a/ in un-

stressed position, and 172 cps lower than that of /a/ in stressed posi-

tion. Based on these data, the author draws the following conclusion:

"The contrast of frequencies puts /a/ in final position in plurals in a

unique situation comparable with the middle frequencies of /e/ and /o/s

together with this articulatory closed quality, palatalization-typical

of other Spanish dialect areas-is minimal here" (p. 12). Guitart (1973)

This Ph.D. dissertation, which is primarily theoretical in

nature, devotes one chapter to a description of the phonetic manifesta-

tions of the writer's native dialect, the Educated Spanish of Havana

(ESH). The author describes ESH as a dialect whose "speakers can be

characterized as university educated and belonging to the middle or

upper-middle class" (p. 40). Guitart states that although he and

other Cubans who speak ESH, whose speech the author's observations are

based an, have lived in the United States for at least five years, "none
shows in his Snanish any noticeable phonological influence from English"

(p. 69). He notes that all these speakers are more or less fluent in
In discussing Guitart (1973), only those observations of the

author that are unique or that differ with other sources will be dis-

cussed, i.e., such items as seseo, yejsmo, or deletion of /d/ in

past-participles will not be mentioned. With this limitation in mind,

Guitart (1973) presents the following data relevant to the pronuncia-
tion of ESH:
1. [s] is a dorsoalveolar articulation, with the tongue convex.

2. /s/ is frequently aspirated in syllable-final position, and

is frequently deleted in utterance-final position, but /s/ is never

realized as [] within an utterance, e.g., esto [ghto]; hoy es lunes

[oy6hlune]; estas cosas [ehtahkosa], but never [ehtak6sa] or Cetak6sa]
"although these are common in less prestigious dialects" (p. 43).

3. There is no voicing assimilation of /s/ (whenever /s/ is not
realized as [h)) when it is followed by a C[voicej segment, e.g., asma
[Isma] and not [~azma].

4. All utterance-final nasals-are realized as []2.

5. [h] is the surface realization of the voiceless velar frica-
tive [x] found in other dialects.
6. ESH does not show word-final /r/-/l/ confusion, "which on the

other hand, is a common phenomenon in the speech of poor Cuban blacks

and of other lower socio-economic groups" (p. 43).

7. "The vowels of Cuban Spanish seem to be more open than those
of other Spanish dialects" (p. 44).
8. Frequently e / v in normal con-
-stress/ 10 C-y / -high

versation within a word and across morpheme boundaries, but only in some
words and never in others, e.g., pasear [pasysrr, but le6n [le6nj and
never [ly6n].
9. More generally /o s / > V ,hig e.g.,
tress -> igh
toalla [twaya]; cohete [kwete].
10. /N/ in ESH appears as [3] before dental, alveolar, alveo-
palatal, and velar consonants, while a velarized labial nasal occurs
before labial and labio-dental consonants, e.g., un boleto [ugboleto];
un francis [uifranses1; un domingo [u3domifgo]; un senor [ugselor]; M
chiste [u6ste]; un caballo [Cu aSyo]. This surface manifestation of
nasals before other consonants is, of course, very different from the
standard language, where nasals before other consonants are homorganic.
Guitart observes that "one of the characteristics of ESH is a strong
tendency toward velarization" (p. 47).
11. /a/ -: [O] /__Y_ e.g., una islita [unihlita].
12. The stop and continuant allophones of /b d g/ are in
free-variation in ESH, whereas in other dialect areas of Spanish they
are in complementary distribution.
13. In careful speech in ESH, the voiceless-voiced/stop-con-
tinuant opposition are neutralized in closed syllable-final position,
both within and across morpheme boundaries, i.e., /B/ may appear as
[p b f; /D/ may appear as [t d ; /G/ may appear as
[k g- ], /B D G/ are labial, dental, and velar archiphonemes

respectively, e.g., absolute may appear as [apsolito] [absolGtoJ l

[absolutto]; apto may appear as [apto] [abto] [a$to].
14. In rapid speech: in ESH, the tendency toward velarization be-
comes active, and only /G/ appears. Guitart provides the following
absolute [aksolfotol
eclipsar [ekliLsar]

[tnico [aniko]
admitir [a mitir]

15. The neutralization of the liquids /r/ and /1/ occurs in ESH
in closed syllable position before certain consonants in the following
manner in colloquial speech
a. The systematic phonemes /1/, an apico-dental lateral, and

/r/, an apico-alveolar flap, are realized as such in prevocalic, inter-
vocalic, and utterance-final positions. /1/ and /r/ "maintain their
identity before other liquids, dorso-alveolar [s1, and alveo-palatal
and palatal obstruents," with both /r/ anid /1/" being dorso alveolar
before [s], and /1/ assimilating its point of articulation to the
consonant that follows but remaining a lateral (pp. 50-51).
b. Both /r/ and /1/ become unreleased, voiced dorsoalveolar
stops before dentals and alveolars other than [s].
c. Dental and alveolar consonants (except [sj) following /r/
or /1/ are affected as follows ,
(1) Apico dental segments ([t ]) are realized as dorso-
palatoalveolar, both being realized always as a stop, e.g., arde
{[fd de-; artete ; [[te; saldo [sad o]; alto [Cadtol (following Guitart,
[A 1 symbolizes an unreleased voiced dorsopalatoalveolar stop).

(2) The apico-alveolar nasal is realized as a dorso-
palatoalveolar nasal, e.g., care [kd neJ.
d. Before labial and velar obstruents, /r/ and /1/ assimilate
in point and manner of articulation to these consonants, e.g., el pomo

Ceb'p6moj; el verde [eb'bed-de]; ser padre [seb'epL re; ser bueno
Cseb'bwenoj]; el caso [eg'kaso]; el RolDe [eg'gob'pe]; orgaidea
[og'kddya]; organizer [og'ganisgr]. As shown by the above examples
given by Guitart, in each case /r/ and /1/ is realized as an unreleased
voiced stop, and if the following velar or labial segment is voiced it
is realized as a stop rather than as a continuant.
e. When /r/ or /1/ occur before [m S f hJ, true gemination
occurs, e.g., el medio [eeminyol; el Same [eannme]; ser mejor

[semmehorl; comer !ame [kom~Ennme1; el frente [effrente); ser fino
Cs[ffino1; el jarnn [ehham 1; dar jan6n [dghham6nl.
16. A velar glide is commonly substituted for [t] after a back
vowel, e.g., Cuba [kkGwaj; la boca [law6ka]; sobre [sowre].
17. In rapid speech, the sequence [Fwj may be frequently reduced
to [w], e.g., abuela [awelal.
2.2.2,4 Lamb (1968)
This dissertation contains 126 pages of text, which is followed by
thirty pages of a sample transcription of the interview made with one of
Lamb's informants used in this study.
The data and findings presented in Lamb (1968) are based on inter-
views with thirty informants, all residents of the United States at the
time he did this study, ranging in age from 21 to 54 years, most (11) in
the age range of 40-48 years, the median age of Lamb's informants being
40 years; their average age 39 years. The range of residence in the

United States of Lamb's informants was from three months to seven years,

their average being three and one-third years. Lamb's informants seem to

encompass a good sampling of the average white middle socio-economic and

educational class of Havana Cubans.9

Chapter One of Lamb (1968) gives introductory information as to

the intention and scope of this study, along with a description of the

sections of Havana where Lamb's informants formerly resided. Contrary

to Bartos' (1970) findings showing that labio-dental [vJ was used in

Cuban-Spanish for the grapheme v, especially in word-initial and inter-

vocalic position, Lamb (1968) states that the labio-dental pronunciation

"is done indiscriminately whether the spelling requires a b or a v,

which certainly in a chronic sense reaffirms, contrary to the intention,

the current indistinguishability of these two signs" (p. 8).

In Chapter Two, Lamb gives some historical background information

on Cuba, and he then discusses all previous work done on Cuban Spanish,

with, of course, emphasis on those studies relating to Cuban Spanish

phonology. Lamb cites the following six studies as having to do with

Cuban phonology or pronunciation (p. 13):

1. Almendros 1958

2. Dihigo 1916

3. Espinosa 1935

4. Isbagescu 1965

5. Montori 1916

6. Olmstead 1954

Almendros (1958), Espinosa (1935), and Isbagescu (1968W(which con-

tains the 1965 article) are discussed in section 2.2.1 of this study.

The remaining three articles, Dihigo (1916), Montori (1916), and

Olmstead (1954) are only briefly mentioned in section, because

of a basic agreement with Lamb that they contain only a small amount of

phonetic material, and are of little interest from the point of view of
a phonological study.

Chapter Three of Lamb (1968) discusses the methodology used by the

author in his data collecting process, and then lists important statis-

tical data about his thirty informants.

In Chapter Four Lamb (1968) discusses the segmental and non-seg-

mental taxonomic phonemes of Cuban Spanish. He lists 18 non-syllabic

and five syllabic segmental taxonomic phonemes. Lamb carefully dis-

cusses each of the 23 segmental taxonomic phonemes, giving a statement

as to the articulatory correlates of each, and the distribution of the

different allophone(s) of each.

Contrary to Bartos (1970), Lamb states that the taxonomic

phoneme /d/ is "usually dental" (p. 46),

In discussing /ptk/, Lamb makes no mention of these segments ever

becoming voiced in intervocalic position in rapid speech, as was first

reported in Saciuk (1974).10 Curiously, however, in the sample trans-

cription of the interview with one of his informants, Lamb transcribes

this informant's response to a drawing from FLPI, 'lena haciendo

candela,' as [I ia asytnio jandelal (p. 129), instead of the expected

[lea asygn/o kandela], clearly showing a /k/.--. [C / V_ V. Even
more interestingly, later in Chapter Four, Lamb (1968, p. 79) actually

states that "/k/ was not heard to have an alternate [j], as stated in

the Isbagescu article."

Lamb (1968) lists /x/ as a taxonomic phoneme of Cuban Spanish, and

neither in this chapternor in the following chapter on phonetic

variation, does he list Ch] as an allophone of /x/. The Cubans I have

been associated with, some having lived in this country for many years,

others being recent arrivals from Cuba, simply do not use [x] as a

regular part of their phonetic inventory, but rather, they use [hJ. I

find no reason, for NCS at least, to include /x/ as a systematic pho-

neme. To do so, would require a rule of absolute neutralization of the

type /x/--4 [h] in all environments, a type of rule not espoused any

longer by many generative phonologists.II

Lamb's discussion of the nasal phonemes /m n i/ does not agree

with what is currently observed in NCS. This disparity may reflect a

difference in the phonologies of MCS and Cuban Spanish, but most likely,

it was already taking place in Cuban Spanish, when Lamb collected his

data, presumably around 1966-67. Lamb states that [j] is, "a velar

articulation occurring before velar non-syllabics and in final position

whenever, across word boundaries, it anticipates either a following

velar non-syllabic or a tonic syllabic" (p. 49). A discussion of the

surface level manifestations of [q] in MCS is given in section 3.3 of

the present study.

The remainder of Chapter Three of Lamb (1968) discusses the

structure of the syllable in Cuban Spanish, i.e., in what possible com-

binations, and where in the syllable, these 23 taxonomic phonemes may

occur, and the prosodic features of Cuban Spanish. In his discussion

of sequences of the same syllabic (pp. 63-64), Lamb makes the rather

curious statement that "rapid speech, however, on occasion causes a re-

duction to one syllabic" [emphasis minel. Apparently, from this, we

are to presume that in rapid speech, it is not only possible, but more

frequent, to have two contiguous identical vowels pronounced as two

units, rather than as one. This type of presumption, however, can only
be erroneous, since the simplification of two identical contiguous
vowels is a normal process of all unaffected speech, and the only
possibility in rapid speech.
In his discussion of phonetic variations in Chapter Five, Lamb
discusses "the phonetic variation observed in the entire corpus of the

data, of the performances of all the informants including individual
differences . given in generally descending order of frequency
observed" (p. 72).
In his discussion of the occurrence of the fricative allophone

[]J, Lamb mentions the occurrence of j] in his data in the environ-
ments h E syllabic] and m# r, rather than the standard [b]
usually found in the latter of these environments. Lamb reports, con-

trary to Bartos (1970) and Istbaescu's (1968) findings, "no instance of
the reverse phenomenons occlusion of the normally fricative [Jl."
Lamb (1968) devotes, of course, a great deal of discussion to the
allophones of Cuban Spanish found in most of the literature, and these
will not be repeated here.
Lamb (p. 76) mentions that word-final d, in addition to its usual
allophone [f], is sometimes realized phonetically as [hl, e.g., pared

[pargh], and also as [0], although he attributes the latter to being
"more limited to the low cultural level."
Contrary to what is common in MCS, Lamb found that the [Cs allo-

phone of /c/ "occurred predominately with women informants" (p. 79).
In NCS, [t] as a surface manifestation of systematic /c/ is common in
both sexes. The phonetic realizations of the systematic /c/ are dis-
cussed in section 3.7 of the present study.

In his discussion of the allophones (pp. 81-82), [s], [h], and [C
of /s/, Lamb's data support the findings in Hammond (1973) that /s/---
[Cf /.tj within a word, contrary to the findings of others, e.g.,
Bertot (1969, p. 9). Contrary to the findings of Hammond (1973),
Olmstead (1954), and Rosario (1956), which all report a compensatory
lengthening of vowels before /s/-. [CO $ within a word, Lamb re-
ports that he did not "observe any consistent effect on the remaining
syllabic. . with no alternation of the final syllabic's quality"

(p. 83). Lamb, however, is referring to vowels before deleted /s/ in
word-final position. He apparently did not consider distinguishing
those vowels before deleted /s/ in syllable-final position within a
word from those in word-final position.
Lamb's findings (p. 84) concur with the findings of this study on
MCS that voicing assimilation of /s/ before voiced consonants does not
occur in Cuban Spanish or MCS.
Lamb (1968) reports the following surface manifestation of
systematic /r/:
1. Evidence of devoicing in word-final position, e.g., vivir

2. ~r] in utterance-final position in emphatic speech, although
"this was a rare occurrence among my informants" (p. 88), e.g., comer

3. [hI before another non-syllabic, usually when this
non-syllabic is a stop, e.g., poraue [pohke].
4. Before another non-syllabic, there occurred less frequently
than [h], "an assimilation to the following non-syllabic resulting in a
double non-syllabic" (p. 88), e.g., acuerdo [akweddo]; cerca [sikkal;
cane [Cnne].

5. The coarticulation of [1] and [r], giving a lateral flap

[ti, e.g., tuerto [twatto].
6. A substitution of [1] for [r], e.g., garganta [galanta]

7. When /t/ or /d/ follows /r/, the liquid may be articulated as
a partial flap, with the tongue striking, and then being held against
the alveolar ridge, continuing then to the articulation of the follow-
ing /t/ or /d/, e.g., cerdo [sed o1; s.rta rsat.aV (the raised /f/ and

/A/ represent the partial flap, and as shown in the transcription, the
following non-syllabic is slightly backed, from dental to alveolar

8. /r/ may appear as [1] in absolute final position. The pho-
neme /A/ in Lamb's data may sometimes appear less tense than the normal
trill, sometimes approaching a voiceless apico-alveolar fricative in

rapid speech.
Lamb reports (p. 91) that "assimilation of /n/ before bilabials

/p b/ is observed with good consistency" in the case of the indefinite
article un.
/n/ in final position may be weakly articulated, or deleted,
nasalizing the preceding vowel, e.g., un dedo [Ct/o].
Lamb reports that /1/ may alternate with /r/, e.g., sol [sor],
and as in the case of /r/, it may assimilate to a following non-syllabic,
e.g., olvido [obbflol; Alberto [abbettol. He also found final /1/ to be
weakened, and infrequently deleted, e.g., tal [tTl.
In his discussion of vocalic segments, Lamb discusses eleven
allophones of the standard five vowel phonemess /i e a o u/. Each of
them has an open and close allophone; /a/ has three allophonic mani-
festations, [a] a central phone, [a] a more-forward prepalatal

articulation and [a] a more-back, velar allophone. Lamb gives the fol-
lowing distribution of these vocalic allophones:
1. /i/ -- [~1 /_/m n/ in pretonic position, and sometimes be-
fore /r/, except after /p t/, or after another syllabic. [ii occurs
2. //-. [u] before a non-syllabic within a closed syllable,
e.g., ixpU [pilpol, "but even more noticeably when in the same environ-
ment and between secondary and primary stresses conductor [kondukt6r]"
(p. 95). [u) occurs elsewhere.
3. /e/ -' [e] before /r 7/; before any [-syllabicj segment in a
closed syllable; in atonic positions; in /i e/ diphthongs before
/n m 1 t/; before another syllabicic] segment / [t], e.g., cohete
Ckoite]; before /d/ in a closed syllable, e.g., sed [sTj.
[e] and [el occur in free-variation in certain environments before
/s/ and /t/, although Lamb's explanation of these environments is some-
what less than lucid (p. 96). [el occurs elsewhere.
4. /o/ [Col / /y E S/; in all closed syllables except
before /1/. [ol occurs elsewhere.
Lamb states the final (unstressed ?) /o/ can be weak, and change
timbre to [u], most frequently after palatals, e.g., serrucho [seMIul];
cello [kweyu]j; cuchillo [kuffyu].
5. /a/-4 [a /Cal y cI/ / /n/ $ C /___ [-sylla-
bic] N, e.g., asco [9skoj.
/a/-. [a] / [il; and in post-tonic position, e.g., rasa
/a/-) [a] generally elsewhere.

In Chapter Six of Lamb (1968), the author attempts to correlate

the findings from his data with other linguistically related areas in

the Caribbean, comparing his data with that found in other published

dialect studies of Caribbean Spanish.

The final chapter of Lamb (1968) gives a summary of his findings,

and his conclusions based on these data.

Lamb (1968) has been discussed at some length because of the

abundance of data it includes. Although some of Lamb's findings do not

concur with my observations, or the observations of others on Cuban

Spanish, his work does provide a valuable source of data, particularly

since he used thirty informants. One potential problem with Lamb's

findings is that, of his thirty informants, their average length of

residence in this country was three and one-third years. Although it

is fairly certain that Lamb's informants who resided in the United

States for six months or less (four) should have been representative

speakers of the Spanish spoken on the island of Cuba, it is an empiri-

cal question whether the other 26 informants (whose average residence

in this country is 3.9 years) really are representative of the Spanish

spoken in Cuba, i.e., that their Spanish has not been influenced by

other variables in their environment. It would be interesting to

correlate the data of the informants who have lived in the United States

for five years or more years (14) with that of the informants who have

resided here one year or less (10). Sosa (1974)

This recent study explores many different areas of the diachronic

and synchronic aspects of the phonology of Cuban Spanish. One chapter

of Sosa's dissertation deals partially with the present-day

pronunciation of Cuban Spanish. A large portion of Sosa's Chapter 4 is

devoted to developing a phonological model of Cuban Spanish. Throughout

this chapter, Sosa makes various observations about the pronunciation of

Cuban Spanish, which are relevant to the present study. It is perhaps

worth noting that this study is quite difficult to read for anyone

untrained in current stratificational theory, as it is written within

that framework. The following are my interpretations of some of Sosa's

observations about Cuban Spanish pronunciations

1. The Spanish of the province of Oriente is distinguished from

that of the other provinces by its intonation.

2. /h/ is heard in Cuban Spanish in place of /x/ of the standard


3. The first syllable of carnaval may appear in any of the fol-

lowing ways in Cuban Spanish [karl; [ka]?; Ekall; [Ia]; [k91]; Ckag];

[C1n]; [kan]; [kah]; Cgfi)]; [kas].
4. Vowels in Cuban Spanish do not necessarily appear phoneti-

cally as more open, as reported by many other writers, but they may, on

the contrary, show the opposite tendency, i.e., toward being more

closed. The phonetic shape a vowel takes "depends on its position and

on the conditions that accompany its production," i.e., [Istress], the

neighboring segments, etc. Sosa adds that "not only does there exist in

Cuba the i that tends to be like . but also that tends to be

like I and o that tends toward u" (p. 127).

5. Sosa suggests that the relative open/closed quality of a
vowel in Cuban Spanish is determined by what amounts to a type of

transderivational constraint; i.e., a given vowel, e.g., /e/ may appear

as standard [e], may become more raised, appearing as [i], or lowered,

appearing as [ e; we therefore have a range for /e/, from [eC to [i],

with the ideal surface output being [el. Sosa states that the choice of

one of these "different phonetic solutions depends on the degree

('grado') of phonetic liberty given by the redundancies present"

(p. 128). Sosa suggests that when there is an absence of semantic or

morphological information available rendering the phonetic output am-

biguous, the more ideal vowel tends to appear; on the other hand, when

there is redundant information supplied, there is a tendency for the

more ideal surface variants to appear (p. 128). This is, of course, an

interesting hypothesis, but unfortunately, Sosa offers no acoustic or

perceptual evidence to support this theory.12

6. Unstressed vowels may, particularly in word-final syllables,

disappear completely, and also this entire word-final syllable may

disappear. If this syllable final vowel is deleted, "there generally

occurs a lengthening of the vowel in the preceding syllable. Fre-

quently the 'omitted' vowel is, in reality, expressed as a voiceless

vowell casa can be produced as [kgisa"](p. 130).

7. Two non-identical contiguous vowels may be reduced to one

vowel, or to one syllabic nucleus,

8. Two contiguous identical vowels undergo vowel reduction,

hosee length will tend to be greater than that of the simple vowel, but

subject to variation according to specific cases" (p. 136).

9. In common words, the tendency to shift stress and then undergo

diphthongization, present in most Spanish dialects, does not occur in

Cuban Spanish, e.g.:


Most Spanish dialects




b wl





Sosa specifically excludes learned words, e.g., perfodo, austrfaco,

cardfaco, from the above process, because "These words are diphthongized,

i.e., stress shift to the penultimate syllable, in Cuba as in the rest

of the Spanish world, with the learned form--without the diphthong--

existing in very reduced circles" (p. 141).

10. Verbs ending in -ear /ear/ appear as [jarj on the surface,

which has been reported in many different studies on Cuban Spanish, but

Sosa specifically excludes the verbs mear and apear from this process

(p. 143).

11. Neutralization of liquids, especially in syllable-final
position, is very common in Cuban Spanish, e.g., auedar [kerr] .

[kejgl [kealt; Pner [poner] Eponel7 point] ; cerca [srka] [sgka];
syllable-final liquids may also be deleted.

Cuban Spanish

stress shift


stress shift



stress shift


12. Syllable-final liquids may become nasalized, especially when

there is a nasal segment present in an adjacent syllable, e.g., salcochar

Csaqko rj; carnival Ckajnagrl].
13. In the speech of Whites from the lowest class, and in the

traditional speech of Cuban Blacks, liquids are most frequently deleted
in preconsonantal and final positions.
14. In the speech of peasants and Cuban Blacks, /1/ may also ap-
pear as [y] in preconsonantal and final positions, e.g., porue Cpoykel;
ser [sey].
15. A tendency exists to insert a vowel between an obstruent and
a liquid when these form a consonant cluster, e.g., prado Cparuu];
yarruma Cyaur5ma'. This inserted vowel can also be pronounced with
full vocalic quality, creating semantic confusion between lexical items,
e.g., ~arado--rado, both being pronounced as [parCo!.
16. Sosa suggests that /r/ --> [kr / # _, as reported in

Isbagescu (1968a), may be more correctly described as [kf], with the
word-initial /r/, "being converted into an affricate with initial pro-
nounced occlusion which gives the impression of the corresponding
stop", e.g., rato [KVto( (p. 151).
17. In everyday speech, /n/ followed by the semivowel [y] be-
comes palatal C[n, and the [y] is deleted, e.g., demonio CdIminU];
Tonia [Tina]; boniato [bUitU].
18. Nasal segments tend to be substituted for all other con-
sonants in post-vocalic position, especially if there is another nasal
consonant present in an fdl wing syllable, e.g., aluno [ang~nU];
desde Cdinde]; bye bye Cbamby7; a nasal segment may be added at the
end of open syllables, e.g., negro [nnegru,; irritado [inrit~Ut ; or a

vowel in an open syllable may simply become nasalized, e.g., satisfaccirn

19. In word-initial position, a vowel may be deleted, and "can
result in a syllabic nasal", e.g., entonces [VtinsI] (p. 153).
20. Nasals are assimilated to the following obstruent, as in
standard Spanish, e.g., canto [ICntU1; concho [kncU]; angulo [a)gilUj];
convidar [kombir,]; but a nasal before /r/ appears as C11, and not as
alveolar [ni, e.g., honra [ogra].13
21. Unlike standard Spanish, however, a nasal followed by a
fricative consonant, or a nasal consonant in word-final position fol-
lowed by a vowel-initial word within the same breath-group, or a nasal
at the end of a breath group, appears phonetically as [1j, i.e.

[+ cont]
/n/ C q I

e.g., cansada kaqjsg]; son hadas [s6z1aa]; en agua [erwa].
22. /n/--4 [q] before the grapheme h, when prefix-final /n/ is
followed by a vowel, "perhaps because of orthographic influence, or as
a morphemic consequence," e.g., anhelo [ae6lo]; inhumano Eigumanul; but
inamovible [inamowflej (p. 155).

23. V N-' V (N) /__ [-syllabicI e.g., done [Zdej;
son altos [s3A1toh] [ sonaltohl.
24. The phonetic manifestation of /b d g/ as C[continuant] oc-
curs in Cuban Spanish as prescribed by the "standard language," i.e.,
they become [Ccont] after pauses, after nasals, and after /1/ in the
case of [dJ, only in careful speech, e.g., dedo [d6;o]; falda [faldaj]
bobo bb~o]; ga o [go]. In normal speech in Cuban Spanish, the
Scontinuant] and [-continuant] surface manifestations of /b d g/ are in

free-variation, except that the [-continuantJ forms "tend to predominate
in non-intervocalic environments" (p. 157).
25. /b d g/ may become semi-vowels in intervocalic position or
before liquids, e.g., tabla [tawla1; cobra [k6wraJ; Dobre [p6wrej;
Eabana [awanaJ; coda [kSwa]; Cuba [Ckwa] (p. 160).
26. In emphatic speech, the nasal [L7 tends to be substituted for
other nasals, rather than the nasal being homorganic to the following
obstruent, e.g., conde [k6Sde].
27. [C and [/] may be deleted in rapid speech in intervocalic
position, e.g., nada [Cna]; a buscar [awhkarj. [$ X ] in syllable-
final position "is assimilated to the consonant that follows it, is
deleted, or even-in emphatic speech--becomes the corresponding [voice-
less] stop," e.g., subdirector [suirekt6rJ; obtuso Cott5so]; verdad
[beddli] usted [uhte] vs. [supdirekt6r]; [optuso); [berdt]j; [uhtet]
(p. 161).
28. [fj may be deleted when it follows [hi, e.g., las gomas
[lahbmaj; las gallinas [lahayfnal.
29. /e/ -4 [] I/ / _/ [-ar3, e.g., chochear [colr];
catcher [kegri].
30. [S] and ['c are free-variants of systematic /c/, e.g.,
macho [m~sU]; chico [9koj.
31. [k] may become [Y] in intervocalic environments, e.g., la
cosa [laisa]; vaqueta [Ca/ta1; las camas [la4ina].
32. /f/ is realized as [f], i.e., a voiceless bilabial frica-
tive, before [rJ, [1], or a vowel, e.g., cofre [k6/ri; flor [161];
fino [#nu..

33. /s/ appears as Ch] or [Co / #7 in normal speech, and not
as C[s. /s/ only appears as Csi in absolute final position or in pre-
consonantal position in affected speech or in cases where semantic
ambiguity may exist.

34. /s/ /i/ [v] c [-J [0J [vI
1 2 3 1 2 3
e.g., asociaci6n Casoasgin]l comision Ckomilsn].

35. "Orthographic f may be aspirated in Cuba in some words--the
same thing that occurs with h and s," e.g., fumar ChumnrJ; fuerte

[hwrtlj; huvuL o [huyiyu ; hablar [yalr]; hala hal1r] (p. 168).

36. /s/ may appear as [h] in syllable-initial position, e.g.,
decimo K[dhimo]; pese C[phej (p. 169). If, however, a word-boundary

intervenes, the above process does not take place.14

37. /s/ may also be assimilated to a following consonant, e.g.,
hasta [~ttaj; busca [blkka1. Again no data are offered to support this

type of assumption. Without actual physical measurements, it is
doubtful that this type of assumption can be taken too seriously. Is

/s/ actually assimilating to the following obstruent, or is /s/ merely
being deleted?

38. When preconsonantal /s/ is preceded by a word-initial vowel,
this word-initial vowel may also be deleted, e.g., hasta [ta]; estg

Besides the observations of Sosa mentioned above, there are many
other statements given by the author in his Chapter Four regarding the
pronunciation of Spanish on the island of Cuba. The thirty- eight items
discussed above are by no means exhaustive of Sosa's study, but merely

those topics felt to be most important or relevant to the present study,
or those topics not mentioned in most other studies previously done on
the phonetics or phonology of Cuban Spanish. Vallejo-Claros (1970)

This study is primarily socio-linguistic in nature, and it dis-
cusses the distribution of four surface realizations of /r/, two of /?/,

and four of /s/ in terms of their stratification in different geographic
areas and socio-economic levels in Cuba. As it was not the author's

purpose to present data concerning the pronunciation of Cuban Spanish,
except for the surface manifestations of these three systematic phonemes,

the information in Vallejo-Claros (1970) relevant to the present study
is very limited.

Vallejo-Claros' data was gathered during the years 1968, 1969, and

a short period of time during 1970. His body of data consisted of 400

hours of recordings from 750 informants, whose ages varied from twenty
to sixty-five years, with an equal proportion of male and female in-

formants. The informants used in his study were all recent arrivals on

the "Freedom Flights" from Cuba, and the interviews took place in the

Miami, Florida area. (see and above).

Vallejo-Claros considered the following phonetic manifestations of

systematic /r/s
1. [r], a voiced alveolar flap (R-l)
2. E], a voiced alveolar fricative liquid (R-2)

3. [1J, a voiced lateral (R-3)
4. EfB (~-4)

These surface representations were studied in the following environ-
ments: /__ e.g., /amar/i /V V, e.g., /arado/; /C_ V, e.g.,

/prado/; /V C, e.g., /korte/.

The stratification of these four phonetic variables, as well as
the phonetic variables studied for /s/ and /f/, were analyzed in terms

of these three socio-cultural levels,

1. upper class (superior S-1)
2. middle class (media S-2)

3. lower class (inferior s-3)
and in terms of the following three geographic areas
1. Western Cuba (G-l), the provinces of Pinar del Rfo, La
Habana, and Matanzas;
2. Central Cuba (G-2), the provinces of Las Villas and Camaguey;
3. Eastern Cuba (G-3), the province of Oriente.
For purposes of the present study, the following data on the
phonetic manifestations of systematic /r/ are presented

1. [r] occurred an average of 8Q% of the time in Vallejo-Claros'
data [Cl 1l%; [11 2-; [] 1V. In terms of socio-cultural classes, the
following distribution is given (all percentages are averages):
a. upper class [r] 88

El lr,
Col 10%

b. middle class Cr] 7%9

W ^18

c. lower class [r] 73
[Ci 23%

[11 3%

With respect to geographic distribution, the following averages for the
surface realizations of /r/ are noted

a. Western Cuba

b. Central Cuba

c. Eastern Cuba

[rI 66;

[rJ 83&
Ir 192
C11 Xr

PI 93%

CE1 ?%


Vallejo-Claros included the following surface realizations of
systematic /I/ in his study
[E--a voiced alveolar trill (R-l)
[-]-a lengthened alveolar liquid fricative (R-2)
These phonetic manifestations were studied in the following environ-
ments: /# e.g., /rosa/; / V, e.g., /aos/; /C V, e.g.,
The following information, relevant to the present study, is taken
from Vallejo-Claros (1970):
1. [ ] occurred an average of 94V of the time; [C] 6%. With re-
gard to socio-cultural classes, the following distribution of [i] and
C[l is given (all percentages are averages):
a. upper class [Cr] 97
o" 2

b. middle class [r1 94

c. lower class [r 927S

With respect to geographic stratification of systematic /r/, the follow-
ing is noted
a. Western Cuba [r] 90S

b. Central Cuba [r] 9%

c. Eastern Cuba [Cr 98"S

In Vallejo-Claros (1970), the following surface realizations of
systematic /s/ were considered,
Csi--a voiceless predorsal convex alveolar fricative (S-1)
[s!--a voiceless apico-alveolar fricative (S-2)
[hi--a voiceless pharyngeal fricative (S-3)

[1 (s-4)
The above phonetic manifestations of systematic /s/ were studied in the
following environments: /# e.g., /sinko/; / _, e.g., /ihos/;
S V, ge.g., /pasar/; /C. V, e.g., /bolsiyo/; /V C, e.g.,
/poste/; /C C, e.g., /instruir/.
The following information, relevant to the present study, taken
from Vallejo-Claros' study is presented:
1. [sI occurred an average of 525 of the time; [s] 6,; [h] 35-;

[C 7/. With regard to socio-cultural classes, the following distribu-
tion is given for the surface realizations of /s/ (all percentages are

a. upper class [s] 56%

[hi 35
[J 4%
b. middle class [s] 52%

[hi 35

c. lower class [s] 4%

Es'l 6
[hi 361

With respect to the geographic stratification of the surface level
manifestations of systematic /s/, the following is noted (again all
percentages are averages):
a. Western Cuba [s] 5C0

Chi 3r

b. Central Cuba [si 495

[hi 34
[co] n
c. Eastern Cuba [s] 58

[hi 36%
[ol 5

It seems obvious from the above data that the much more frequent

aspiration of /s/ in Western Cuba, mentioned by other authors, is not

supported by Vallejo-Claros' findings, unless one wishes to make the

claim that there is some significant difference in 394, 32%, and 3&6, or

at least enough of a difference to claim that aspiration of /s/ is much

more frequent in Western Cuba.

Besides the data given above, there is a great abundance of

further socio-linguistic data presented in Vallejo-Claros (1970) which

have been omitted here, as they are beyond the scope of the present


2.2.3 Papers Presented on Cuban Spanish

In this section, one paper, the only one to my knowledge given on

Cuban Spanish which deals with phonological or phonetic data, is dis-

cussed. It is, of course, highly possible that other papers exist,

which have been overlooked, since there is no regular published listing

of papers presented at all linguistic conferences. Guitart (1974)

This paper is largely theoretical in nature, and therefore pre-

sents a minimum of phonetic detail. Guitart does, however, provide the

following information relative to pronunciation in three dialects of

Havana, Cuba:

1. "In one Havana dialect spoken by uneducated Whites" systematic

/s/ and /f/ both are phonetically realized as hJ / C in rapid

speech; e.g., esto [ehtoj; difteria [dihterya ](p. 2). Although I have

not observed /f/ -> [h / C with any degree of regularity, I have,

nevertheless, observed /s/ [h] / C as a frequent occurrence in

my informants from Havana, all of whom are educated Whites. It would

appear then, that /s/ [h7 / C, in MCS at least, is not a
phenomenon limited to uneducated Whites. This matter will be discussed
more fully in section 3.3 of this study.
2. In the same Havana dialect discussed above, /r/ -> [hi

/I +sonoranti, e.g., care [kahne,; Carlos [kghlo]. Again,
Guitart attributes this phenomenon to uneducated Whites from Havana.
Among my informants from Havana, with university education, the pro-
nunciation of the word Carlos regularly appears in the following ways:
[k1rloJ] [I~hlol [k lo]. Thus, it appears from my informants from
Havana, that this phenomenon is not limited to uneducated Whites in MCS.
3. In "Black Cuban" speech, /s/ may appear as [ / C, e.g.,
esto [rtoi. Again, I must object not to Guitart's observation of this
process, but rather to his relegation of /s/ -` [C /_ C_ to the
speech of Black Cubans. Among my informants, /s/ -- [] / C ap-
peared with regularity, at least in rapid speech. This phenomenon ap-
pears in the speech of my informants, all educated Whites, who come from
all six of the Cuban provinces.
4. In the speech of uneducated Whites from Havana geminationn
occurs in the realization of non-strident obstruents in syllable-final
position in rapid speech," e.g., doctor [do't6r]; acto and arto
[ta to]; abdomen [ad'd6menl (p. 5). The geminated non-strident ob-
struent is realized as unreleased, and the following segment, if
/b d g/, is realized as a stop, rather than as a continuant.
5. In the Educated Spanish of Havana (ESH), in rapid speech,
systematic /r/ and /1/ are neutralized in syllable-final position.
Their surface manifestation depends on the feature [coronall of the
following segment: (a) Liquids appearing before C[coioned segments

become geminated, with the geminated liquid being realized as unre-
leased. Guitart states that "in addition, if the second consonant is
/b d g y/ the stop allophone occurs" (p. 5). This statement is diffi-
cult to follow, since /d/ and /y/ are both [+coronal]. Guitart gives
the following examples
ala and ama [Imma]
el bobo [e'b6bo]
ser bobo [seb'b6So]
el aato [eg"'gto]
e1 llamS [e5amS]
(b) Liquids appearing before [-iced oro segments are assimilated, but
the feature [+voiced] of the systematic liquid is retained, e.g., e
radre [eb'pthre]; el caso [eg'I~so]. (c) Systematic liquids occurring
before [-coronal] segments, except /s/, are realized as an unreleased
interior segment; the [+coronal] segment following the liquid is
realized also as [-anterior], e.g., el dormingo [ed domirgo]; rimer
domingo [primed domfgo] ([d] is used by Guitart to represent
non-anterior /d/). (d) Systematic liquids occurring before /s/ are
realized as a flap (represented as [Ci) before other [* coronal] seg-
ments, e.g., el socio [eji'syo]; ser socio [sers6syo], with the /s/
being realized as [-anterior]. Guitart notes that the realization of
/s/ as [-anterior] "alters its timbre. .. with [s] sounding more
similar to English [s] than to English [s]" (p. 6).
2.2.4 Unpublished Studies on Cuban Spanish
In this section, two studies done by Tracy Terrell are discussed.
Although these two studies are, at this time, unpublished, they are,
nevertheless, of interest because they not only raise some important

questions relative to phonological theory, but they also provide a

corpus of data upon which Terrell bases some of his later published

studies, e.g., Terrell (1975a, b, d), and (1976a). Terrell (1974a)

In this study Terrell presents data concerning the retention,

aspiration and deletion of systematic /s/ in the Spanish of Havana.

His data are taken from tape recordings made during the "Estudio co-

ardinado de la norma lingUistica culta de las principles ciudades de

Iberoam6rica y de la Peninsula Ibgrica"; this same group of taped in-

terviews serves as a data source for other studies by Terrell, for

Haden and Matluck (1973 and 1974), and for Vallejo-Claros (1970). The

author's purpose in this study is not, however, that of providing a

large body of data, but rather to propose a linguistic explanation,

based primarily on weakening processes, of s deletion and s aspiration

as.it occurs in different environments in this dialect.

Terrell (197La) presents data relative to s retention, s aspira-

tion and s deletion in syllable-final position within a word, and

word-final position within a breath-group based on the following


1. retention--the [s] and [zi surface manifestations of /s/;

2. s aspiration and s deletion--the [h], [hP or [Cf surface

forms of /s/, i.e., deletion and aspiration of /s/ is considered as one

group, and is contrasted as a whole with A retention.

Terrell (1974a) presents the following data, of interest to the

present study:

1. In syllable-final position within a word, /s/ is retained 3%

of the time and aspirated or deleted 97% of the time.

2. In word-final position, /s/ is retained, i.e., [Cs or [z]

33% of the time and either aspirated or deleted 67V of the time.

3. In word-final position, /s/ is retained 5% of the time before
consonants, and is aspirated or deleted 95% of the time in this environ-


4. /s/ is retained 30% of the time before vowels and is

aspirated or deleted 7Q of the time when a vowel is the next immediate


5. /s/ is retained 62, of the time before a pause and is
aspirated or deleted 38% of the time in this environment.

As interesting as the above data from Terrell (1974a) may be, it

is not possible to compare his percentages with those given for the

surface forms of systematic /s/ in Chapter Three of the present study

because Terrell analyzes these data based on syllable-final position

within a word and word-final position, while the present study dis-

cusses the different surface manifestations of /s/ based on syllable-

final position within a word, word-final position within a breath-group

and absolute-final position. Also, Terrell presents his percentages

based on two groupings--[s] and [z1 versus [hj, [fi and [f], while the

present study presents its percentages based on the categories--s re-

tention, [s] versus s aspiration, [hi and [h]; versus s deletion, [$. Terrell (1974b)

Although the author's purpose in this study is to propose an

explanation for why s aspiration and s deletion occur in different

environments, based on phonological and grammatical constraints, there

is, nevertheless, some data presented which is relevant to the present

study. This study uses the same data source as Terrell (1974a).

In the data and in Terrell (1974b), of 7149 occurrences of /s/,

17% appeared phonetically as [s], 635 as [hI [], and 2Q0 as [C.

The data given in Tables XXI, XXII, and XXIV in Chapter Three of

the present study show that of 3415 occurrences of /s/, the surface

form Cs] occurred at a percentage rate of 4.7 [h] occurred at a rate

of 46.24, and [C] at a rate of 49.1L.

Comparing Terrell's data with the data on systematic /s/ in

Chapter Three of the present study, it can be seen that in both studies

[sj is the least frequent occurring surface manifestation of /s/, al-
though Terrell reports an occurrence rate of 17? while in this study

only 4.7, is reported. The difference in these two percentages could

be due to several different variables; the most likely of these

variables is probably the interview situation. Terrell's data is based

on a more formal interview environment, while the data used in this

study were obtained from a totally non-structured, social environment.

It would seem likely that the more formal the interview situation, the

higher the percentage of s retention.

Terrell (1974b) reports a 63% rate of s aspiration versus a 46.2,

given in the present study; Terrell (1974b) reports a 205 rate of s

deletion versus 49.1V reported in the present study. The differences

between the percentages of s aspiration and s deletion in these two

studies may be due, in each case, to the authors* exact interpretation

of s deletion. Terrell considers, for example, a lengthened vowel re-

placing a vowel followed by /s/ as a case of aspiration and not dele-

tion (personal communication), whereas this is considered to be dele-

tion in the present study. This difference alone would greatly affect

the reported percentages.

In syllable-final environments, Terrell (1974b) reports [sI oc-
curring only 2V of the time, [h]*C[i 9 7, and []j, 13.

In the present study, in syllable-final environments (Table XXIV),

[sl appears at a rate of 9.45, [h at a rate of 70.,%, and [CJ, 20.3,.
Although the specific percentages differ, both studies show that in
syllable-final environments, [h] is by far the most frequently occurring

surface manifestation of /s/.

Terrell (1974b) reports the following percentages for the dif-
ferent surface forms of /s/ in word-final environments within a

breath-group: Preconsonantal Prevocalic

C[s] 1Of

[h 7% 50%
[n1 2 3 3W-
Terrell (1974b) presents a great deal of data on s retention,
aspiration and deletion in relation to different grammatical cate-

gories, in an effort to support the hypothesis of the study. Some of

his conclusions are (1974b, pp. 29-30);

1. Deletion is a process of word-final environments; aspiration
occurs both in syllable-final and word-final environments.

2. There are three relevant phonological environments for s
deletion and s aspiration: preconsonantal, prevocalic, and prepausal.

3. The rule of s aspiration occurs most frequently in preconson-
tal environments and least frequently in prepausal environments, while

the percentages of application of the different environments of the
rule of s deletion are inversely related, i.e., A deletion occurs most

frequently in prepausal position and least frequently in preconsonantal


4. s aspiration is largely unaffected by grammatical constraints,

with the exception of determiners in pre-vocalic position.

5. "s deletion, on the other hand, is primarily correlated with

morphological classes and grammatical functions and is constrained by

functional considerations" (1974b, pp. 29-30).

6. Deletion is suppressed in prenominal modifiers and in the

verb form es.

7. In Cuban Spanish, there is no structural necessity for an

expanded seven-vowel system, as proposed in many previous studies on

different Spanish dialects (see Hammond, 1973, pp. 1-5).

2,3 Miami-Cuban Spanish

Because the MCS dialect area is quite new in comparison to other

American-Spanish dialect areas, it is understandable that there has not

been an overabundance of research done on it yet. Almost all of the

research done on ICS that has found its way into print, or has been

presented in the form of papers read at linguistic conferences, has

been carried out within the state of Florida, and more specifically, at

either of two universities within the State University System of

Florida. Because almost all of this research has been done in the

form of masters theses, doctoral dissertations, or papers read at con-

ferences, it is not surprising that it has not received very wide dis-

tribution. Because this material has not been readily available to

others thus far, the content of each of these seven studies on MCS

known to me will be briefly discussed in the remainder of this section.

2.3.1 Published Materials on .CS

As far as can be determined, Resnick and Hammond (1975) is the

only published article that deals specifically with MCS. This article

summarizes the experiment and findings done in Hammond (1973). This

experiment deals with the surface manifestations CsJ and [C] of systema-

tic /s/ in MCS. Since this article and Hammond (1973) deal with the
same data, Resnick and Hammond (1975) will not be discussed more fully

here, but rather in section 2.3.2 below, in the discussion of Hammond

2.3.2 Masters and Doctoral Research
Done on 'CS
After extensive bibliographical research, it appears that there

are only two masters theses and, including this study, only two doctoral

dissertations which specifically deal with MCS. The two masters theses

are Hammond (1973), and Fernandez (1973).

Hammond (1973) is an experimental study which attempted to verify

the hypothesis often-repeated in the literature on Spanish phonology15

that a compensatory change in the vowel formant structure takes place

(in those dialects of Spanish where /s/ -- [C0) when this vowel pre-

cedes the surface manifestation [oJ of systematic /s/. Although no

consistent significant change was observed in the formant structure of

vowels appearing before /s/ -> [CO / $ in MCS, there was, however,

a consistent observable (both acoustically and perceptually) compensa-

tory vowel lengthening which took place in the vowel immediately pre-

ceding deletions of systematic /s/ in syllable-final environments within

a word. Thus, although the morphemic and/or semantic distinctions

carried by systematic /s/ were neutralized in this environment, the

additional vowel length served to compensate for this neutralization.

Therefore, after the application of the late rapid speech rule of

syllable-final s deletion, a minimal pair, e.g., busaue 'look for'

[bGskel--buaue 'ship' [buke] should become neutralized, both lexical

items appearing as [bike]. However, because of the process of compen-

satory vowel lengthening, they are still distinguished, after the

application of Rule I, by differences of vowel length, i.e., as busque

[Cbs kej-buqaue [bakel. This type of data found in Hammond (1973) will
be discussed more fully in section 3.5 of this study, and the interest-

ing theoretical implications that arise from this process in MCS will

be discussed fully in section 4.3 of the present study.

Fernandez (1973) is a study of lexical and syntactic borrowings

from English which are used in MCS. Most of Fernandez's study concerns

lexical borrowings, such as mopear, from the English verb 'to mop,'

frequently used in MCS instead of standard Cuban Spanish trapear

(pp. 41-42) or flonquear (un curso), from the English 'to flunk (a

course),' instead of the standard Spanish suspender (un curso) (p. 39).

Fernandez gives the phonetic representations for each of these borrow-

ings. He also discusses some syntactic borrowings from English that

are found in NCS. One example Fernandez gives is the placement of ad-

verbials between auxiliary verbs and participles (pp. 66-67), e.g.,

Juan ha rapidamente salido 'John has quickly left,' based on this type

of syntactic pattern found in English. The type of syntactic surface

structure oft


allowable in surface structure in English and NCS is impossible in

standard Spanish, in which *Juan ha r4pidanente salido is ungrammatical,

and must be expressed as: (1) Juan ha salido rapidamente; (2) fP.Dida-

mente Juan ha salido; or (3) Juan r~pidamente ha salido. Fernandez

(1973) also gives some background sociological information on the ECS

dialect area.

Besides this study, the only other doctoral dissertation being

done on KCS is Bjarkman (forthcoming) which is primarily a theoretical

study, using MCS as a data source which may be used to test some of the

claims made by the theory of Natural Phonology proposed by Stampe.

Bjarkman (forthcoming) applies the concepts proposed by Stampean

Natural Phonology to the investigation of linguistic borrowing. Al-

though Bjarkman's study deals with loan-word phonology in MCS, it makes

no claim to have provided a complete description of all phonological

processes involved in linguistic borrowing in MCS, or to have accounted

for all such borrowings. This study uses loan-word phonology of MCS as

a means to an end, not as an end in itself.

2.3.3 Papers Read on MCS

As far as can be determined, after careful research into the

matter, there have been three papers read on the MCS dialect, all within

the last two years.

Resnick and Hammond (1973) discussed some of the findings and data

of Hammond (1973), and dealt primarily with some of the acoustic and

perceptual impact of s deletion in the MCS dialect. This paper compared

the formant structure of vowels before and after the deletion of a

following /s/. The phenomenon of compensatory vowel lengthening before

deleted /s/ within a word, but not word-finally, in MCS was discussed.

Saciuk (1974) discussed the voicing of intervocalic voiceless

stops in the rapid speech of some speakers of MCS who originally came

from the Cuban province of La Habana. This study showed that this

voicing process applying to all three /p t k/ takes place only in very

rapid speech, and apparently never occurs if the rate of speech is de-

creased.16 This study also compared this synchronic phenomenon of

intervocalic voicing of voiceless stops in YCS with the well-known

historical process that took place in Old Spanish, in which intervocalic

voiceless stops became voiced.

Hammond and Resnick (1974) discussed the history and development

of the hypothesis that there is a change in the formant structure of

vowels occurring before surface manifestations of deleted /s/, i.e.,

that these vowels become more open in this environment, as a compensa-

tory mechanism for the loss of the morphological and semantic informa-

tion carried by /s/. It was shown that no such vowel formant restruc-

turing takes place in YCS, and that in word-final position when /s/

-b [] in NCS, the singular-plural distinctions carried by /s/ in

nouns and adjectives, and the second vs. third-person distinctions

carried by /s/ in Spanish verbs are lost, and that these distinctions

can only be adduced from linguistic context.

2.4 Other Spanish Dialect Areas

In this section, several topics, found in other Spanish dialect

areas, relevant to phonetic or phonological processes of Cuban Spanish

will be discussed. The dialect areas described in the articles dis-

cussed in this section are, for the most part, dialects that share many

phonological similarities with Cuban Spanish, e.g., Puerto Rican

Spanish, PorteSo Spanish, Andalusian Spanish, etc.

One of the articles discussed in this section, del Rosario

(1956), is at times dealing with Cuban Spanish, but under the heading of

Caribbean Spanish. For this reason, this article will be given a more

detailed discussion than the other studies mentioned in this section.

Only some of the relevant portions of the remaining seven studies

discussed in this section will be presented very briefly here, as a

general discussion of their contents is beyond the scope of this study.

2.4.1 Cassano (1972)

Cassano attributes the phonetic process of the occurrence of [Js

as a surface variant of systematic /c/ of Porteio Spanish to French

influence. This process is also widely found in Cuban Spanish and in

MCS, for which a claim of French influence would be difficult to


Cassano also discusses the appearance of [1] in word-final posi-

tion in the Spanish of Argentina, also as a result of French influence,

e.g., campp] as a surface realization of champa'a, based on French

champagne ['Spn"l. The appearance of the nasal [o' in word-final posi-

tion, along with other possible phonetic solutions, will be discussed in

sections 4.6 and 4.7 of the present study.

A characteristic of Argentinian Spanish discussed by Cassano,

which is not related to French influence, is the development of an

eight-vowel phonemic system. Cassano states that "the loss of the

plural marker (sl in other dialects of Argentine Spanish (and in the

dialects of other Spanish American countries as well) has resulted in

new functional opposition based on vowel quality," e.g., va u], vas

[Cbl (back /a/); dio [dy,! Dios [dy]l (p. 176). Cassano, however,

offers no empirical evidence to support this contention, a claim often

made about those dialects of Spanish in which syllable-final /s/ may

optionally appear as [].

2.4,2 del Rosario (1956)

In his chapter entitled "The Antilles" de Rosario devotes five

pages to a discussion of the pronunciation of the Spanish in Cuba,

Puerto Rico, and the Dominican Republic. Some of the author's observa-

tions about the Spanish of Cuba, or about these three countries in

general are:

1. A vowel system containing seven phonemes /a e e i o o u/.

The seven-vowel system "resulting from the change _sh the aspirate

Eh] opens the preceding vowel, and on eliminating it [the [h]] the open
[eI and [o], slightly lengthened, remain, which can occur also with the
a" (p. 83). Del Rosario offers the following examples of this pheno-
menon: peso [pso.]--pesos [peso]; pie [py j-pes [pygel; come [k6mel]-

comes [k6mel; dio [dyl--Dios [dyol. In this discussion, however, del

Rosario is referring to the Spanish of the Antilles in general, so it
is difficult to determine whether the author is making the same specific

claim about Cuban Spanish. At any rate, he offers no acoustic or per-
ceptual data to support his claim of a seven-vowel phonemic system (or
later an eight-vowel system).
2. The frontal or predorsal /s/ occurs more frequently than the


3. /h/ occurs in place of the /x/ of the standard language,
e.g., Jrajia [p.ha.
4. /s/ -, [hI in syllable-final position, e.g., pescar [pehkr].

5. /n/ [~] /._ _j, e.g., an [pa"1; jam6n [ham6i].
6. "Total assimilation Cof /r/] to a following consonant is heard
at times in the Antillean region, but is much more frequent in the

western part of Cuba" (p. 82). Del Rosario offers the following
examples observed in Havana: cuerpo [kweppol; puerta [pwettal;
reform [ref6mmaj.
2.4.3 Honsa (1965)
In discussing the taxonomic phonemic inventories of six dialects

of Argentinian Spanish, Honsa lists an eight-vowel system for the dia-
lects he refers to as Standard Buenos Aires; Colloquial Buenos Aires;

Lunfardo; West and North Argentinian; provinces of Corrientes and

Misioness /a e ee i o o u/. Without offering any supporting evidence,

Honsa states that when the allophone [hi of /s/ is deleted at the end of
syllables, "an additional phonemic contrast of vowels steps in to dif-

ferentiate meaning. This contrast is neutralized in all other

syllables" (p. 277).

2,4.4 Hooper (1973)

In this study, Hooper, who is arguing for a phonological model

based on the tenents of Venneman's Natural Generative Phonology, uses

data from Saporta (1965), Mondejar (1970), and Alonso et al. (1950),

which have to do with open vs. closed distinctions in vowels. Along

with the theory espoused by Cassano (1972), Honsa (1965), and many

others, there is claimed to be an opening (laxing) of vowels in sylla-

bles checked by /s/ in those dialects where /s/ may appear as CC

/ $. This vowel opening or laxing serves as a compensatory me-
chanism to preserve the morphological or semantic information carried

by /s/. There is no acoustic or perceptual evidence presented by any
of these authors to support the above theory, and Hooper, as have many

others, apparently accepted this open-closed vowel distinction as

valid. As previously mentioned, in an attempt to prove the validity of

such an open-close distinction in vowels before /s/, in one Spanish

dialect where /s/ may optionally appear as [] / $, neither acoustic

nor perceptual evidence could be found to support such a theory

(Hammond, 1973). While the fact that such a morphophonemic change, as

suggested by an open-closed vowel inventory, was disproved for one dia-

lect where /s/ may appear as [] (YCS), it remains an empirical question

whether such a morphophonemic change could or could not occur in other

dialects where /s/ deletion occurs. Until such claims can be supported

in some manner by data from these dialects, such claims must be, at

best, regarded with a certain amount of skepticism. Fortunately for

Hooper, the correctness of these claims will not appreciably weaken any

of her arguments.'

2,4.5 M ondejar (1970)

Although this study directly concerns itself only with the

Andalusian verbal system, it does, nevertheless, touch on some items of

pronunciation relevant to Cuban Spanish. The Spanish of Andalusia and

the Spanish of the greater Caribbean area, of course, share many

phonological and phonetic processes. The Spanish of Andalusia, like

Cuban Spanish, is a dialect where /s/ may be realized also as [h] or

CC at the phonetic level. Mondejar makes the interesting claim,

although he unfortunately presents no supporting evidence (in terms of

spectrographic evidence or perceptual studies), that in Western Anda-

lusla, when /s/ appears phonetically as C] in verb endings, "the

speaker uses other differentiating elements for the lack of phoneni-

cized vocalic openness to show the already mentioned opposition: the

personal pronouns" (p. 40). On the other hand, Mondejar claims for

Eastern Andalusta a ten-vowel phonemic system, with each vowel of the

five-vowel system of standard Spanish splitting into an open vs. closed

phoneme /A a e e ii o u u/ (p. 38). Mondejar states that "the

loss of -s produced a non-differentiation in the system, a fissure, a

weak point. The phonemicized vowel openness reestablishes internal

order to avoid such a weak point in the system" (p. 40).

Mondejar lists the following phonetic manifestations of word-

final -Z in Andalusian Spanish: [r 1 h I r], with the first three

surface realizations being the most common.17

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs