Thought and World in Wittgenstein's Tractatus Logico-Philosophicus

Material Information

Thought and World in Wittgenstein's Tractatus Logico-Philosophicus
Smith, Andrew
Publication Date:


Subjects / Keywords:
Aboutness ( jstor )
Existence ( jstor )
Metaphysics ( jstor )
Nominalism ( jstor )
Nonsense ( jstor )
Ontology ( jstor )
Signification ( jstor )
Symbols ( jstor )
Thought ( jstor )
Truth ( jstor )
Tractatus logico-philosophicus (Wittgenstein, Ludwig)
Wittgenstein, Ludwig, 1889-1951
Undergraduate Honors Thesis


This paper examines Ludwig Wittgenstein's book, Tractatus Logico-Philosophicus. Specifically, I attempt to determine whether Wittgenstein is committed to a certain kind of metaphysics, what I call "thought-world metaphysics." Thought-world metaphysics is a metaphysical inquiry that seeks metaphysical explanations for the relationship between things in the world and features of thought. It seems that Wittgenstein is committed to what I call "thought-world metaphysical realism" - the thesis that the existence of certain categories of the world explains certain features of thought. I address the work of Marie McGinn, who claims that Wittgenstein is not committed to this thesis. I argue against her interpretation, and in doing so I compare parts of her anti-metaphysical interpretation to the anti-metaphysical thought of Rudolf Carnap in order to illuminate her interpretation and to show why Wittgenstein should not be interpreted in her way. After this, I claim that Wittgenstein is not committed to thought-world metaphysical realism because we have no reason to think he is not also committed to the opposite thesis of thought-world metaphysical irrealism; namely, that the features of thought explain the existence of entities in the world. I then conclude with what this means for the place of metaphysics in the text. ( en )
General Note:
Awarded Bachelor of Arts; Graduated May 7, 2013 cum laude. Major: Linguistics
General Note:
College/School: College of Liberal Arts and Sciences
General Note:
Advisor: Gene Witmer

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Andrew Smith. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.


This item is only available as the following downloads:

Full Text


1 Thought and World in Wittgen Tractatus Logico Philosophicus By Andrew Smith Submitted for consideration for honors in Philosophy to the Undergraduate Committee of the Philosophy Department at the University of Florida, April 10, 2013. Advised by Dr. D. Gene Witmer, Associate Professor and Chair of the Philosophy Department at the University of Florida


2 Introduction Ludwig Wittgenstein Tractatus Logico Philosophicus covers just about every conceivable philosophical topic: metaphysics, logic, language, ethics, God scientific explanation, probability, the foundations of mathematics, and philosophy itself Although Tractatus Logico Philosophicus (hereafter TLP ) is a sho rt book, Wittgenstein believed that it solved all the problems of philosophy or at least explained why all of them exist: The book deals with the problems of philosophy, and shows, I believe, that the reason why these problems are posed is that the logic of our language is misunderstood. The whole sense of the book mi g ht be summed up in the following words: what can be said at all can be said clearly, and what we cannot talk about we must pass over in silence. 1 This paper will focus on the status of metaphysics in TLP. Specifically, this paper is an attempt to figure out whether or not Wittgenstein is committed to a particular kind of metaphysics. The early sections of the book c ontain metaphysical statements, consisting in statements about the nature of the world and its constituents. Later, the book moves into describing the fundamental nature of thought, language, and logic. Wittgenstein thin ks that propositions of thought and langua ge mean what they do insofar as they are pictures of rea lity. They picture reality insofar as the y pictures sha re a form with reality, which is to say that the pictures must have as many parts as the state of affairs being pictured. One gets the sense in r eading the passages that discuss the picturing relation between thought and the world that the relationship between thought and the world is of central importance to the status of metaphysics in the book. It seems that this relationship discussed in TLP wo uld commit Wittgenstein to the existence of entities in order to explain the way that thought works. 1 Rout ledge and Kegan Paul, Ltd: 1963, p. 3


3 It is the above idea that I will pursue in this paper. It seems Wittgenstein is committed to a certain kind of metaphysics dealing with the relationship be tween thought and the world, what I will call thought world metaphysics. Specifically, Wittgenstein seems committed to what I call thought world metaphysical realism. A philosopher is a thought world metaphysical realist if she believes that certain catego ries of entities exist, and believes that they exist because they explain the features of thought. I will spell out a TLP. The question then becomes whether Wittgenstein i s actually committed to the premises of such an argument. While there is a large body secondary literature on TLP, I Elucidating the Tractatus because I find some challenges to interpreting Wittgenste in as holding such an argument in her book. I will argue that a couple ways of phrasing her arguments against reading metaphysics into TLP fall flat. However, I believe that one way of characterizing her challenge is important. Specifically, it is the idea that the concepts in those concepts which correspond to the categories of the world that the prima facie realist argument states Wittgenstein is committed to In order to flesh out this idea, I will compare th is line of thought to the later work of the philosopher Rudo lf Carnap, giving my own interpretation of his paper language that refer to traditional ontological categories can be explicated as variable styles in linguistic frameworks. Because Carnap believes that the form that a linguistic framework takes is a matter of preferential choice and not a matter of fact, this means that we cannot be committed to any ent ities purely on the basis of the truth of certain


4 follows trivially from how a linguistic framework is set up. Thus, the explanation for our es of signs which refer to en tities is the fact that we prefer to use that category Considering the fact that Wittgenstein also represents traditional ontological categories in his conceptual script via variable styles, the question arises whether Wittgen stein is also committed to explaining categories of symbols in his conceptual script by way of preferential choice rather than by way of the existence of entities in the world. I will argue that this is not the case, as Wittgenstein believes that the categ ories of thought which are in question in the prima facie realist argument names and propositions have to be present in any adequate conceptual script. Thus, the prima facie I will argue that it does no t hold up to a challenge to a different challenge, one not found in I will argue that the prima facie realist argument as I will state it could be rephrased as an argument that concludes that the existence of cate gories of entities is explained by the features of thought. This rephrasing of the argument as an argument that concludes with what I will call a thought world metaphysical irrealist thesis calls into question the idea that we can draw out of Wittgenstein any theses which commit him to the idea that we accept the existence of entities to explain the features of thought. This is because, in the irrealist version of the argument, the explanation goes the other way around. Thus, I conclude that Wittgenstein is not committed to thought world metaphysics in the sense that he believes there is a metaphysical explanation to be had with respect to the relationship between thought and the world, although it is probably


5 true that he is committed to the existence of su ch entities in a weaker sense that I will specify. I divide the thesis into sections. I will start the paper in Section 1 by summarizing TLP, and in Section 2 I will note the debate surrounding what to make of the fact that Wittgenstein decla res it to be n onsense and why I will ignore this debate for the purposes of this thesis. In Section 3 I explain in Section 4 I draw out the prima facie realist argument that one seems to find in TLP. I begin my discussion of challenges I can find to the prima facie realist interpretation. In Section 7 I compare the most important challenge from McGinn to Rudolf Carnap, and Section 8 I argue why Wittg enstein cannot be compared to Wittgenstein in this res pect. In Section 9, I argue that there are reasons to think Wittgenstein is not a thought world metaphysical realist after all. I will conclude with a suggestion for what this means for understanding th e status of metaphysics in TLP. § 1 A Brief Summary of TLP I will now summarize TLP. There are many important points in the text I will skip over for the purposes of brevity in summarizing the text. I am attempting only to give a general picture o f what i s going on in the book. There will be other claims made in the text that are not apart of my summary but will be noted later on in the paper. TLP is not written in standard philosophical prose. Rather, it is grouped into numbered propositions. T he propositions are numbered 1 through 7, with all the other propositions numbered with decimal points following any one of these numbers.


6 cimal numbers assigned to the individual propositions indicate the logical importance of the propositions, the stress laid on them in my exposition. The propositions. n .1, n .2, n .3, etc. are comments on proposition no. n ; the propositions n.m 1, n.m 2, etc. are comments on the proposition n.m, citations of TLP in this paper will be followed in parentheses by the number of the The seven main propositions of TLP are the follo wing: ( 1) The world is all that is the case ( 2) What is the case a face is the existence of states of affairs ( 3) A logical picture of facts is a thought. ( 4) A thought is a proposition with sense ( 5) A proposition is a truth function of elementary p ropositions. (An elementary proposition is a truth function of it self). (6) The general form of a truth function is [ p N ( )] 2 This is the general form of a proposition. (7) What we cannot speak about we must pass over in silence. We can gather the general flow of the text from these propositions. Wittgenstein first gives a general description of the world and the states of affairs that make it up. Then, Wittgenstein moves to the relationship between thought and facts, and how thoughts are logical p ictures of facts. Logical pictures of facts are propositions, and every proposition is a truth function of elementary propositions. Thus, Wittgenstein relates the general account of pictures and their relationship with reality to his view on the nature of logic. We then are given the general form for constructing any proposition whatsoever in the formal logic that Wittgenstein advances. The book ends with a poetic imperative commanding us to be silent about what we cannot speak about. What we cannot speak a detail. 2 What I represent as a dash to the left is actually a vertical bar over the symbols in TLP.


7 Let us look further at the propositions that support these seven propositions. t the world a fact states of affairs that Reality is Wittgenstein d efines it existence and non (2.06) Wittgenstein then claims that there are things called pictures non The elements of a picture are made up of objects that are representatives of objects (2.13 2.131). the elements of a picture are related to one another in a deter minate way represents that Pictures have a certain form, must have in common with reality, in order to be able to depict it correctly or incorrectly Although a pictorial form is (2.172). A picture depi and non sense with reality c


8 the possibility of the situation of which it is the thought. What is thinkable is possible Wittgenstein relates thou ghts to propositions and what he calls propositional signs Wittgenstein argues that a propositio nal sign is a kind of picture as discussed before, in For Wittgenstein, language admits of logical analysis into pro positions with simple signs. the propositional sign correspond to the object of the thought. I call such elements 2 3.201). The simple e sense; only in the nexus There is a philosophical purpose to analyzing propositions. Wittgenstein distinguishes between a sign and a symbol a sign and the same sign (written or spoken, etc.) can be common to two different symbols in which case they will signify in same word has different modes of signification


9 out as his task as dispelling these errors caused by such confusions by using a sign cludes [errors] by not using the same sign for different symbols and by not using in a superficially similar way signs that have different modes of signification: that is to say, a sign language that is governed by logical grammar (3.3 25). understand a proposition, I know the situation that it represents. And I understan d the communicates a situation to us, and so it must be essentially connected with the situation. sition is a be exactly as many distinguishable parts as in the situation that it represents The two 04). Wittgenstein makes a distinction between what a proposition shows and what a proposition says, and how this relates to the logical form of propositions shows its sense. A proposition shows how things stand if it is true. And it says th at they show can be shown, cannot This has consequences for what Wittgenste in thinks


10 sphere of natural science. It must set limits to what can be thought; and, i n doing so, to 4.114). The rest of the book goes about giving the basics for how such a logical clarification of thoughts would work. In doing so, Wittgenstein attempts to demonstrate what sorts of things are merely shown but not said. The propositions falling under 4.12 describ e how it is that certain ontological concepts things having do with the nature of the world. These particular passages are important to the content of this paper, bu t will be discussed in detail further below. In such a project of the logical clarific ation of thoughts, we arrive at not just any sort of proposition, but what Wittgenstein c elementary proposition fits the way Wittgen stein described propositions before in An If something is an elementary proposition, states of affairs being independent of one another the truth or falsity of each elementary proposition does not rely upon th e truth and falsity of any other such proposition. disagreement with truth These truth possibiliti es can be expressed by a truth t able. The following truth table is a propositional sign, which Wittgenstein uses at 4.442:


11 p q T T T F T T T F F 3 F F T If we take p and q to be elementary propositions, every proposition can be represented as the result of a logical operation on those propositions, given the possibilities of their joint truth or falsity on each row of the table. proposition is a truth function of The foregoing may seem to be merely a logical point, but Wittgenstein has a philosophical gr possibilities of its truth 01). Wittgenstein then says: truth of a proposition follows from the truth of another proposition if all the truth grounds of the latter are th e truth gro inference can be borne out by setting out the truth possibilities of propositions in truth tables and seeing whether they share the same truth possibilities that make it true. Wittgenstein starts ma king the larger point in 5.13 5.131: When the truth of one proposition follows from the truth of others, we can see this from the structure of propositions. If the truth of one proposition follows from the truth of others, this finds expression in relations in which the forms of the propositions stand to one 3 Wittgenstein leaves this spot blank, but considering that he says that, if put on a vertical line, the table I inserted


1 2 the existence of the propositions. The way in which the truth of all propositions whatsoever are related to oth er propositions involves the internal, formal relationship between propositions. This internal that produces out of it other propositio are the familiar logical operations or constants: negation, disjunction, conjunction, and so on. For Wittgenstein, these operations are fundamentally different from functions. Wittgenstein characterizes elementary propositions as functions of names. Howeve because operations can be defined by one another, and because operations can cancel one The general idea here is that the nature of logic is something which language does not itself say anything abo ut but is rather something that is shown by a logically analyzed language. If all propositions are the result of truth operations upon elementary propositions, and this relationship between propositions stems from the nature of the proposition, it is possible to give a general form of a proposition by giving the operation that co nstructs function is [ p N ( )] As Bertrand Russell p stands for all ato mic [elementary] propositions. stands for any set of propositions. N ( ) stands for the negation of all the propositions making up taking any selection of atomic propositions, negating them all, then taking any selection


13 of propositions now obtained, together with any of the originals and so on 4 This result depends upon the interdefinability of truth operations by way of the Sheffer stroke, which is a logical operation which proves that all logical operations on any two propositions p and q can be obtained from ~p & ~q. It also depends upon universal and existential quantification as the conjunction and disjunction of propositions, respectively. As a bove, we here have a logical idea which is aimed to express a philosophical point; namely, that we can construct all propositions from a general plan which takes as the base elementary propositions independently true of one another, and which constructs al l propositions according to a form which any and all propositions share. This general form of the proposition, in being characterized by an operation and not a function, is not something that can be said, but is shown in propositions when we regiment them in a logical fashion. Throughout propositions 5 and 6, Wittgenstein discusses some consequences of this sort of view for the notions of causation, the a priori, ethics, God, mathematics, and the like. Just before the book ends, in 6.53 6.54 Wittgenstein s ummarize the task of the The correct method in philosophy would really be the following: to say nothing except what can be said, i.e., propositions of natural science i.e. something that has noth ing to do with philosophy and then, whenever someone wanted to say something metaphysical, to demonstrate to him that he had failed to give a meaning to certain signs in his propositions. Although it would not be satisfying to the other person he would not have the feeling that we were teaching him philosophy this method would be the only strictly correct one. My propositions serve as elucidations in the following way: anyone who understands me eventually recognizes them as nonsensical, when he has u sed them as steps to climb up beyond them. (He must, so to speak, throw away the ladder after he has climbed up it.) He must transcend these propositions, and then he will see the world aright. 4 From Page xv in the 1963 edition


14 Wittgenstein is here noting the fact that the book has consisted of propositions which themselves say things which can only be shown. The nature of reality is mirrored in propositions, something that can only be shown. The logical properties of proposition s that enable one to construct the general form of a proposition cannot be spoken about but instead can only be shown. Yet, Wittgenstein uses propositions, propositions that presumably would admit of an analysis into truth functions of elementary propositi ons. However, since such an analysis only admits of that which can be said, the propositions of TLP that describe the logical properties of language and its reflection in the world do not actually say anything. Wittgenstein caps this startling conclusion w ith the final proposition of TLP, § 2 The Problem of Nonsense Thus, the aim of this paper as stated seems not to be able to get off the ground. For Wittgenstein declares that the m etaphysical propositions of 1 and 2 are nonsense. We might then ask: what does it mean for the propositions to be nonsense? We may think like this: the propositions of TLP propositions are ways of ge tting us to see how logic and the world really are, even if what can be said.


15 This way of looking at the book has been fought against of late by some Wittgenstein scholars 5 Cora Diamond says What counts as not chickening out is th en this, roughly: to throw the ladder away is, among other things, to throw away in the end the attempt to take seriously the out is to say that it is not, not really, his vi ew that there are features of reality that cannot be put into words but show themselves. What is his view is that that way of talking may be useful or even for a time essential, but it is in the end to be let go of and honestly taken to be real nonsense, p lain nonsense, which we are not in the end to think of as corresponding to an ineffable truth. Diamond and other philosophers, such as James Conant, are grouped together in a reading of TLP which is supposed to the TLP th at I reading of TLP, rather than dealing with the odd this conclusion itself nonsense, the resolute reader takes Wittgenstein not to b e offering a theory of nonsense of any sort, and instead makes seemingly grandiose philosophical claims only to show that they are as much nonsense as anything we would consider to be nonsense by our ordinary sense of the word. 6 This debate is an important one the above merely scratches its surface However, for the purposes of this paper, I will set it aside. The goal of this paper will be to determine what metaphysical commitments if any Wittgenstein has in TLP. This probably means I am assumin 5 Chapter 6 of The Realistic Spirit, pp. 179 204 6 also in a piece co written by Diamond and Conant Tractatus Resolutely: Reply to Mered ith Williams and Peter Klbel, Max and Weiss, Bernhard, eds. The Lasting Significance of Wittgenstein's Philosophy.


16 in that I am assuming we can draw theses and commitments out of a text that according to them is plain nonsense. Thus, if the resolute reading is the correct reading, the foregoing is pointless. Howe ver, given the fact that many readers of TLP have tried to draw out theses from the book, I am simply following in their footsteps. Not only that, but a reading of TLP which takes seriously the doctrines of the book is a reading with a wider range of influ ence in the history of philosophy. Bertrand Russell and the philosophers of the Vienna Circle were influenced by specific doctrines in the book. Trying to determine commitments of a book as if they pointed to ineffable truths would be to consider a book as many important philosophers have read it. Perhaps these philosophers read the parts of the book that appealed to their philosophical motivations, motivations which the resolute readers tell us Wittgenstein wanted us to discard by concluding the book in sa ying that those appealing parts of the book were in fact nonsense. It is clear, however, that perhaps that book may have some interesting views or commitments to draw out. consider few times in TLP, and it would be an interesting project to see what Wittgenstein might However, I would like to consider what sort of metaphysical commitments he has given a particular conception of metaphysics that I will later explain. The project of drawing out metaphysical commitments in TLP will be to take certain notions of certain kinds of me taphysics that we would encounter in all corners of philosophy With a pre existing


17 knowledge and familiarity with metaphysics, we should be able to read TLP and see things or fail to see things in the text that would meet our criteria for metaphysics. I do not think this implies some sort of anachronism. What it means is that there will be some things Wittgenstein says in TLP that match the terms used in metaphysical debates we are familiar with. Then, we can see whether Wittgenstein is committed to metap hysical doctrines with those terms. This is a fairly normal way of describing how to reconstruct concepts and arguments when examining philosophical texts. The reason this is difficult is that it is difficult to make sure that the concepts and terms we use in our reconstruction of the arguments match reasonably well with the different terms and concepts in the texts in question. It is akin to the problem of translating a foreign language; in that some translations are clear cut, whereas others have wiggle r oom, and still others are a challenge to translate at all. However, these are the necessary pains of doing history of philosophy, as it is the job of the historian of philosophy to try to understand by her own lights what previous thinkers thought by their lights. The previous paragraph about the project of the historian of philosophy is compelling as stage setting for trying to understand, say, Plato. He lived a long time ago, lived in a very different place, and spoke a different language. By contrast, Wittgenstei n published TLP ninety two years ago. It was written in German and translated into English, but that is a much less serious obstacle than Attic Greek, a language written and spoken in a much different time and context than ours compared to the English and German of early twentieth century Europe. These facts should make us feel more comfortable approaching the sort of translation project for TLP.


18 However, there is a particular difficulty applying for this sort of translation project to TLP For one, Wittge nstein is obscure. I feel no need to cite any secondary source to provide evidence for this claim. We probably have a better time understanding Russell and Frege on the whole because they are generally clearer. obscurity is not just in lack of clarity in this or that proposition. It is the fact that TLP is simply not standard philosophical prose. It is an arrangement of statements in the indicative whose arrangement does not necessarily coincide with steps in an argument. It is not indicated that a proposition with a higher value number follows from another different topic. Because of these difficulties, reconstructing arguments is a chore. Howeve r, it seems almo st to be a necessary chore. The bringing of the philosophical terms and concepts we have in order to understand the task seem most pressing for TLP Unlike what one might do with many texts in the history of philosophy, one is often not replacing concepts and terms in prose arguments for TLP but one is rather putting in terms and concepts w hen they are often absent. However, what is the biggest problem with approaching TLP is the obvious one discussed earlier the fact that its author declares it to be nonsense. What point is there to draw out theses if there can be no theses to draw out? This problem is particularly pressing when thinking about the resolute reading. A traditional reading is usually fine with interpreting points in the text, but then add s the addendum that they are nonsense sensu strict o. Again, r ather than try to wade into the aforementioned debate over nonsense in TLP, I will set aside the d ebate for the purp oses of this thesis This is not to say that the


19 debate can be safely ignored. I have discussed the debate for a reason any attempt at a summarizing TLP requires saying something immediately about why Wittgenstein says that it is nonsense. However, as I hope to show, there are important things that can be drawn out of the text if one takes it seriously for a moment. If these things are drawn out, it is still something that any reader of TLP would need to note. If there are fruits to the particular things I have drawn out, then they will be fruits that are not gotten from taking a position in a well worn debate on the status of nonsense in TLP but instead are gotten tments. I suggest that there is value in taking a different tack, the value of which will be borne out as more is written. § 3 Thought World Metaphysics There is one feature of TLP that seems worth noting if someone wanted to capture how it is that metaphysics works in th e text. It seems, roughly speaking, that the properties of language discussed later mirror the properties of world discussed earlier. This mirro ring does not seem coincidental; rather, it seems essential to the nature of language as Wittgenstein discusses it. Take the following propositions: logico in the situation that it represents. The two must possess the same logical (mathematical) Taking these statements at face value, one might think Wittgenstein is committed to some kind of metaphysics here. If the forms of propositions share the form of reality, and propositions must have the same multiplicity as reality, it seems


20 Wittgenstein is attributing certain features to reality. It seems Wittgenstein says that we can certain features of reality off of features of thought. There are other things W ittgenstein says that make it seem we can read certain features of reality off of thought. [ bedeutet ] an object. The object is its A A agreement and disagreement with possibilities of existence and non existence of states of of parts of language has to do with certain categories of things existing outside of language. Given w hat these and related p ropositions say, I believe these propositions contain an important line of inquiry to investigate any possible metaphysical commitments Wittgenstein has with respect to the relationship between thought and language I will clarify th is line of inquiry below. I also believe the specific sort of metaphysical commitment I will investigate is important to understanding the place of Whether one is a resolute reader of TLP or a traditional re ader, it is hard to deny that propositions 1 through 2 (or at least proposition 1) are qualified in some sort of important way later in the text. Either they are completely nonsensical propositions, or they are propositions that point to ineffable metaphys ical truths. However, beyond the issue of their being nonsense, the reason that these metaphysical propositions are important is that they seem to be connected to his views on logic and language later in the book. If one does not pay attention to his views on logic and language, the point of the book is missed. As Wittgenstein says in his not to


21 thought, but to the expression of tho Thus, if his views on logic and language hold a central role, we ought to expect any metaphysical commitments that are worth drawing out to be connected to those views. As has been said, reading off the nature of reality from thought is one kind of metaphysical project. Let us call this project thought world metaphysics Thought world metaphysics is not a particular thesis or set of theses. It is a particular sort of inquiry tha t counts as a kind of metaphysical inquiry. Specifically, it is the investigation into the nature or constituents of reality via the nature of thought and the relationship between thought and reality. ver, t here are known senses for these terms that I hope will suffice in order to characterize the project at the outset. Even so, some c Although I am aiming to characterize thought world metaphysics genera lly, the use of the term for discussing Wittgenstein. For Wittgenstein, a Gedanke pl. Gedanken ) is not used or at least primarily used as we speak of it ; namely, a reflective mental activity As he says: Thoughts are proposition the perceptible sign of a proposit ion (spoken or written, etc.) as a projection of a possible Thus, given the prima facie examples of thought world metaphysics cited above, we can see the relationship between thought and the use of the term s.


22 phil truth and falsity. This bearer need not necessarily be expressed in spoken or written language in every instance of thought a private mental st ate that can be characterized as the bearer of truth and falsity would also count This means that we can carry the idea of thought to the contents of beliefs and knowledge, as well as other propositional a ttitudes. The theses that proceed from thought world metaphysics have a certain form. First, any such thesis is a specific sort of explanation. O ne might develop psychological and biological explanations as to how the world affects our thought. A metaphysi cal explanation is typically a sort of explanation different in kind from any other kind of explanation, an explanation of a fundamental sort different from causal explanations or the like which attempt to explain in some basic way some features of thought and language. A metaphysical explanation is an explanation that purports to show how features of thought or the world are possible where it is presupposed that those features could not exist without some other sort of thing existing. This is at least a necessary explanation could take this form. Distinguishing metaphysical explanation from other forms of explanation perhaps includes, but is not limited to, the fact that metap hysical explanations involve the invocation of entities at a higher level of abstracti on than scientific explanations. Geist exist at a higher level of abstraction than enzymes or neutrinos do. I admit that this is a quite rough c haracterization of the difference, but I believe this is sufficient to at least show t hat we


23 have at least good reasons to think that metaphysical explanations are by and large different from other forms of explanation. I have characterized thought world metaphysics as involving metaphysical explanations about the relationship between thought and the world. What specifically is this sort of relationship? In order to show what I have in mind, it is best here to set out the three general groups that theses o f thought world metaphysics are in. Theses of thought world metaphysics come in three general groups, what I will call realist irrealist and nominalis t A realist thesis states that there exists a category or categories of entities in the world and that the existence of these entities explain (in a metaphysical way, as discussed above) some features of a category or categories of thought An irrealist thesis states that there exists a categor y or categories of enti ties in the world, and that the features of a category or categories of thoug ht explain the existence of the entities A nominalist thesis states that the entities accepted by either the realist or irrealist do not exist, and thus that the explanation for the features of thought is to be found e lsewhere either in other sorts of entities or in some other sort of explanation altogether These theses need not be attempts to explain all the features of all thoughts or the existence of all sorts of entities all at once. Usually, such theses focus on specific features, specific categorie s of thoughts, and specific categories of entities. I also am aware that the terms irrealist contradicting usages in the history of philosophy. However, I believe that the terms have an affinity with how such terms have been used in philosophy for positions that fall in line with thought world metaphysics as I have described it.


24 By features of t content, or structure of thought. A thought world metaphysical thesis may attempt to explain reference, meaning, logical form, syntax, or anything along these lines. Again, I am offering vague and disputable terms, but I believe the idea here is sufficiently clear for the purposes of the paper later of thoughts with features in question. Philosophers are not typically interested in, say, the names in general and the existence of spatiotemporal objects in general. Thus, a thought world me taphysical thesis focuses on some category or other of thoughts and entities in the world. In order to defend usefulness of the concept of thought world metaphysics and to make a bit clearer what I have in mind by it, I will give an example of each of the three general groups of theses that can be found in the history of philosophy. Thought world metaphysical realism is perhaps the easiest of these groups of theses to find, and is perhaps the most intuitively acceptable of the three groups of theses. Hilary notion o ll with my notion of thought world metaphysical th e relation of any 7 supposed to apply to all 7 Proceedings and Addresses of the American Philosophical Association, Vol. 50, No. 6 (Aug. 1977), pp. 483 498


25 8 The idea is akin to how I have stated thought world metaphysical realism in that it says that the thoughts expressed in scientific theories have their content explained by certain features of the world. Putnam does not make the distinction between categories of thought and particular parts of thought that I have insisted on, but it is implied in his description of the model of metaphysical realism that the category of thought he is talking about is the category of the relation of reference is given by the truth cond itional semantics for the language, in the canonical versions of the theory i.e., understanding a term, say, T 1 consists in knowing 9 Thought world metaphysical irrealism is not as easy to find. A good example of t hought world metaphysical irrealism is Hil which is a Reason, Truth, and History. A quote from his Reason, Truth, and Histo ry will suffice: community of users can correspond to particular objects within the conceptual scheme of those users We cut the world into objects when we introduce one or another conceptual scheme of description. Since the objects and the signs are alike internal to the scheme of description, it is possible to say what 10 Thus, the existence of categories of e ntities in the world is explained by the features of thought; namely, the ways in which conceptual schemes are introduced. The example of Putnam shows that 8 p. 484 9 p. 484 10 Reason, Truth, and History, p. 52


26 thought world metaphysical irrealism or another particular conceptual scheme value upon theory relative property of THE WORLD. And the more such sentences there are, the more properties of TH E WORLD will turn out to be theory 11 So in this sense there is no notion of the world working in an irrealist thesis as given by Putnam, but this need not be the case. Thought world metaphysical nominalism closely corresponds to many versions of nominalism as it is usually described. Nominalism typically is the denial of the existence of abstract entities or universals Thought world metaphysical nominalism need not be about abstract entities or universals; however, one particular for m of thought world metaphysical nominalism that one could identify in the history of philosophy is that general or universal terms do not have anything corresponding to them that are themselves general or universal. A good example of someone who adheres to this is Thomas Hobbes Quoting from Leviathan: Of Names, some are Proper, and singular to one onely thing; as Peter, John, This man, this Tree : and some are Common to many things; as Man, Horse, Tree; every of which though but one Name, is nevertheless th e name of divers particular things; in respect of all which together, it is called an Universall; there being nothing the world Universall but Names; for the things named, are everyone of them Individuall and Singular. 12 A s I am construing thought world me taphysical nominalism, the thought world metaphysical nominalist need not deny that nothing corresponds to the features of thought, but rather denies a thesis a realist or irrealist might put forth. Thus, thought 11 12 Part 1, Chapter 4, Paragraph 6


27 world nominalism is perhaps itself not a po sitive position but just is the denial of any sort of particular realist or irrealist position. I believe it is worth noting as one of the three general groups of thought world metaphysical positions because it is important to realize that the denial of th e existence of entities is itself a metaphysical position. The importance of this will appear later when I discuss the views of Rudolf Carnap. in thought world metaphysics. I think this is easier to do given these historical examples relates to all of this discussion. After all, it may seem that thought world metaphysical irrealism has some affin ity to this family of views because, according to irrealism, t he characterization of thought world metaphysics is misleading. The only characterization of about. If one is a Berkeleyan idealist or a phenomenalist, that world will be restricted to minds and the contents of the mind. What is often known in philosophy as realism is any view that opposes this characterization of what thought is about. The point I would like to make here is that these sorts of debates are independent of my notion of though t world met to our specific theories, it does not mean that the world here is some creature of the mind. hat this is not at issue here.


28 § 4 The Prima Facie Realist Argument In the passages of TLP mentioned above o ne is tempted to find realist thesis Recall 4.04: the situation that it represents. The two must possess the same logical (ma thematical) It seems a commitment to entities (distinguishable parts in a situation as in a proposition) is given because that commitment explains the content that propositions hav e. bedeutet ] an object. The object is its A A a category of entities (objects) is given because that commitment explains the content that names have. I will attempt to articulate this prima facie reaction below with an argument with premises supported by propositions from TLP that concludes with a realist thesis. Wittgenstein at v arious points of TLP gives us basic principles for logical notation. TLP contains the rudiments for a conceptual script ( Begriffsschrift ), a system of logical notation that is intended to capture the logical form of thoughts expressed in any language whats oeve r. Wittgenstein calls a constituent of a proposition with sense : A symbol is distin guished from a sign a common to two different symbols, i n which case they will si (3.321). It is a constraint on an acceptable conceptual script that there is a one to one language that


29 excludes [errors] by not u sing the same sign for different symbols and by not using in a to the symbols th at they represent. A written sign qualif ies as a symbol if it has a mode of signification, which a sign has if it is a constituent of a proposition with sense or is itself a proposition with sense In other words, a sign counts as a symbol if it is a meaningful element in a proposition or is its elf a meaningful proposition. I t follows from the above that a difference in signs represents a difference in mode of signification ; in short, if it has a different meaning, broadly construed (I say broadly construed because of issues Let us write this as the first premise: 1) A symbol x is different from another symbol y if and only if x differs from y in sign x is different from another sign y if and only if x differs from y in mode of symbols. There are names and (elementary) propositions a thought can be expressed in such a way that elements of the propositional sign correspond to the (3.2 3.202) function of elementary propositions. (An elementary proposition is a truth There are also relations and functions. Relations are not defined by Wittgenstein but are given in diff erent propositions of TLP in various examples. Functions are related to names and propositions ). I


30 write elementary propositions as functions of names, so that they ha ve the form etc. Or I indicate them by the letters There are also the d ifferent logical constants that represent that a proposition is a truth function of some further propositions. However, these do not qualify as symbols in evident that v, 5.44). Thus, since any elementary proposition can be characterized a s a material function of names 13 and logical constants are not material functions, no logical constant is a constituent of a proposition with sense and thus is not a symbol. If a difference between symbols comes in difference of mode of signification, then we might conclude that a difference in categories of symbols comes in a difference in a category of mode of signification TLP does not contain this thesis anywhere in particular However, it is clear that something like this has to work. The conceptual script will have a and b as distinct letters if they are two different names. However, the conceptual script will have different sorts of letters to distinguish between names and e lementary propositions: the fact that a b, c, etc. are signs for names name and f a ga, fb, g b, etc. are signs for elementary proposi tions is shown by the use of two different types of signs: the former is just one lowercase letter at the start of the alp habet, the latter is an low er case lette ercase letter of the 13 It is implied that material function is another way of talking about a function of names in an elementary world can only determine a form, and not any material properties. For it is only by means of propositions that material properties are represented ties or functions only come in propositions, which matches what is said above about functions being defined as a way of representing an elementary proposition as a function of names.


31 alphabet starting at the beginning. Let us tentatively write this as the second premise in the argument: 2) Members of category of symbols A are different f rom members of category of symbols B if and only if the members A differ from the members of B in category of mode of signification, or category of What could be the difference in a category of mode of signification? Again, Wit tgenstein does not use a phrase like this, bu t what I have in mind is the difference given between the meanings of names and the meanings of propositions in 3.203 and 4.2, respectively. A name is a different kind of symbol from an elementary proposition be cause they have two different sorts of meaning. Now, one must be careful: a name has a Bedeutung and a proposition a Sinn. These are not the same sort of meaning. However, calling both Sinn and Bedeutung their kind of meaning is just a way of getting at the basic point that names and propositions do fundamentally different things. This is different from one name differing fro m another name, which is bec ause they signify different objects, or one proposition differing from another proposition, which is because they represent different states of affairs. It is to say that names differ from propositions because the former signify objects and the latter repr esent states of affairs. If one symbol differs from another if they differ in meaning, and one category of symbol differs from another if they differ in kind of meaning, and yet there are different German terms used for meaning, I do not have a point if t here is not a way to tie these meaning in roughly the same way. I will say that Wittgenstein explains meaning as


32 non different objects. A name differs from a propositio possibility of the existence of a state of affairs. Names and propositions do different things; they are used in different ways in language. In TLP, Wittgenstein thinks the primary use of language, or at least the primary use of language that can be captured in his conceptual script, is this representational use of language. This representational aspect of language is the very basic way in which to explain or thi nk about the content of the constituents of propositions. It is clear that names and propositions are categories of symbols that are characterized by their aboutness. It is debatable whether other categories ought to count as having meaning explained by aboutness. There is debate in the secondary literature in the Tractatus over the status of relations 14 However, if we believe there is a realist thought world metaphysical thesis in TLP, we do not require that all constituents of thought leads to commitments, only that there is at l east one which would be stated in such a realist a thesis. This leads us to th ree other steps in the argument. I present the argument in full below Hereafter, I will refer to this argument as the prima facie realist argument : ( 1) A symbol x is different from another symbol y if and only if x differs from y in ( 2) Members of category of symbol s A are different from members of category of symbols B if and only if the members A differ from the members of B in category ( 3) The meaning of a symbol is explained by its being about something. 14 An overview and a position on this debate, connecting to the question o f whether relations in TLP can be construed as objects in TLP, Synthese (2009): 167: 145


33 ( 4) Thus, the explanation for the fact that a symbol x differs from another symbol y is that x is about something different than y. [ From 1, 3 ] ( 5) Thus, the explanation for the fact that members of a category of symbols A differ fro m members of a category of symbols B i s that the members of A are about some category of entities A* different from the category of entities B* that the members of B are about [From 2, 4 ] This is the path to the prima facie realist thought world metaphysical thesis in TLP Differences in the meanings of categories of symbols are explained (in the metap hysical way) by differences in categories of things of which those symbols are about. This is an explanation of the content of though ts via a commitment to entities in the world, reality, or whatever preferred term for what thoughts are a bout. § 5 McGinn and the Problem of Formal Concepts in TLP Other than resolute readers of TLP that reject drawing out such theses altogether, there are som e readers of TLP who draw theses out of TLP yet consider themselves as being neither traditional nor resolut e readers. An explicit example of such a reading is Elucidating the Tractatus. meaning, which conceives the representing relation as consisting in the existence of a direct link between bits of lan guage (words) and bits of the world (objects) 15 and resolute readings. 16 Wittgenstein in such a way that would challenge premises of the above argument. The basic way in which this chapter is a challenge to the above argument is that it challenges concluding step 5 from 2 and 3. Specifically, we can cast this challenge as claiming that there is an equivocation between 3. The mode of signification that d ifferentiates categories of symbols one from anot her 15 Elucidat ing the Tractatus, p. 3 16 See pp. 5 21


34 cannot be characterized as about ness in the same way that the mode of signification of particular symbols can be. According to McGinn, Wittgenstein does not explain the meaning or mode of signification o f members of categories of symbols by referencing categories of things. As such, we cannot get a commitment to categories of entities. Perhaps we can explain that one sign differs from another in being about different things. However, this explanation is n ot adequate for getting thought world metaphysics, as thought world metaphysics is about categories of entities. It is not a thought world It is, however, a thought world metaphysical explanation to say that the meaning of a name is explained by its referring to an object. Thus, this challenge disputes my attempt to give a single differentiates any sign from another and what differentiates a category of signs from another. The reason McGinn has a challenge of this sort is because of her interpretation of ariables and formal concepts. Objects and states of affairs are Wittgenstein says that an expression presupposes a general form of a proposition that it propositi


35 it designates is due to convention, the class of propositions that results from replacing a co that when we have determined one thing arbitrarily, something else is necessarily th e case. (This derives from the essence Wittgenstein thinks we cannot depict logical form and thus these formal properties in the propositions that represent the relevant relation [formal property] between possible situations expresses itself in language by means of an internal relation [formal property] between the propositions representing h shows how to replace constants by variables to produce a class of propositions governed by a logical form as described in 3.315. pressed in a proposition, but is rather an aspect of signs featured in those propositions (4.126). About these aspects of the signs, Wittgenstein tells us the following: 4.127 The propositional variable signifies the formal concept, and its values signify the objects that fall under the concept. 4.1271 Every variable is the sign for a formal concept. For every variable represents a constant form that all its values possess, and this can be regarded as a formal property of those values. 4.1272 Thus the vari is the proper sign for the pseudo concept object.


36 in conceptual notation by a variable name. pressed by x, y Wherever it is used in a different way, that is as a proper concept word, nonsensical pseudo propositions result. The same applies to the words They all signify formal concepts, and are represented in conceptual notation by variables, not by functions or classes (as Frege and Russell believed). 4.12721 A formal concept is given immediately any object f alling under it is It is clear from these passages that Wittgenstein does not think he is committed to word: it is not a function that figures into elementary pr opositions. It is rather expressed by a variable. The variable represen ts not a material property that holds of some things and do not hold of other things, but rather represents a formal property of certain signs Thus, orld that holds of some things and not other things, § 6 passages. McGinn stresses distinguish the logical role of variables from that of the expressions that are its values. The general proposition xFx does not contain a constituent in which x does indeterminately or a mbiguously what a, in Fa, does determinately. Rather, the propositional sign xFx indicates a logical prototype Fx which is a means of presenting a class of propositions: all the propositions that are th e values of the


37 17 This is significant because variables do not have anything interesting to do construction of a class of propositions that have their logico syntactic properties in common, or which are constructed according to a common logical plan. The rule that a variable expresses does not need to concern itself with what signs mean, but onl y with 18 This insistence on variables being a means for showing how signs signify and not having to do with what is signified is the way in which McGinn argues that Wittgenstein does not have any ontological commitment. As the chapter goes on, McGinn presses this concepts. The thought is something like the following: if what a variable represents is not what the constants mean but is rather expresses the rule for the substitution of constants, we cannot construe Wittgenstein as explaining the meaning of categories i n virtue of a category of entities if there is no connection to the particular entities that particular propositions are about. McGinn believes the confusion about formal concepts Wittgenstein wants to defuse pts are concerned with symbols and their modes of signifying, as this is manifest in their employment in propositions with sense, and not w 19 formal concepts are not functions or classes that hold of something things and not of 17 Elucidating the Tractatus, p. 166 18 Ibid, p. 171 19 Ibid, p. 183


38 what symbols signify, rather th an how symbols sy 20 She says that Wittgen stein wants us to realize that states of affairs in the way that it does, does not characterize, or derive from, the nature of what the expressions of our 21 Thus, f rom the standpoint of the construed sense to categories of symbols because the logical order of language of which the categories are apart are no t explained by things which language is about. If the above considerations truly dispel the move from steps 2 and 3 to step 5, we members of categories of symbols qua members of those categories. insistence that her reading of TLP is anti metaphysical, a clarification of the alternati ve explanation here should not turn out to be a metaphysical on e. The explanation that seems to be offered is that a variab le represents a rule for a formation of a class of propositions Since a variable characterizes those members of signs that fall under differences in variables represent di fferences in those signs that fall under those categories, if the differences between the categories is a difference in rule for formation and substitution of propositions without variables, it is implied that this is not something which depends upon wh at dealing with signs. I take it this explanation by way of rules is why distinctions between categories have to do with how symbols symbolize rather than what the categories symbolize. 20 Ibid, p. 184 21 Ibid.


39 An explanation by way of rules would count as an alternative explanation if a good reason were given why the connection between signs subsumed under a specific between signs The thought here may go somet hing like this: A rule is something normative: it specifies what actions are permissible and impermissible of propositions that have their logico somethi ng that permits or does not permit certain propositions to be constructed. If this rule differentiates categories, it differentiates categories in virtue of what is permitted or not permitted to be done with signs. This explanation refers to the actions of the users of the conceptual script and not the way the world is. This explanation does not strike me as an adequate alternative. For one, we are Furthermore, even if we had some notion of logico syntactic property, the fact that rules may differentiate categories does not mean that the sorts of things the categories are about could not also differentiate categories. Why it is that certain propositions form a c lass that a variable for that class represents and is a variable for which we may substitute constants could be explained by the nature of the sort of things that members of those is that it is not that way. However, merely emphasizing that rules are involved does not make this point something further is needed. The fact that this explanation appeals to rules do es not mean that we can avoid a thought world metaphysical commitment that would also do the explanatory work.


40 There are two things I see that could serve as that further thing. One thing is t he not as of yet clarified notion of logico serves as an explanation differe nt from an explanation referen cing categories of entities. My only guess for any weight that this phrase would hold is an emphasis on the distinction between syntax and semantics. Roughly speaking, syntax deals with rules for well formed expressions while semantics deals with rules th at apply truth, falsity, and meaning to those well formed expressions. For example, syntactical constraints in formal logic make it the case that p & q is a well formed formula, while ~p~q value of false to ~p if and only if p is true, and prohibit assigning the truth v alue of false to ~p if and only if p is false. As far as I can see, an alternative explanation that emphasized syntax o ver semantics would be one that explained the categories of thought as being formal. Returning to the argument above, the emphasis on syntax over semantics would nts for which we can replace a kind of variable from constants for which we can replace a different kind of variable is not due to the meaning of those kinds of variables because in an important sense they do not have a meaning. If differences between cate gories are purely syntactical, this means that what distinguishes a name and a proposition is the different rules for placement in well formed expressions that go verns them. So, one way a pr oposition differs from a name is that a proposition can be the argument of an operation such as negation or any of the logical constants where as a name cannot. These differ ences are syntactical in nature, and only syntactical in nature.


41 The problem with this view is that it is hard to s ee what to make of the idea that the differences are only syntactical in nature. Of course, there are obvious syntactic about formal concepts, what prevents us from say ing that an important difference between names and propositions is that a name means an object and that the sense of a proposition is the possibility of the existence and non existence of the state of affairs that it represents? What prevents us from givin g a basic semantic characterization of the difference between categories on top of the syntactic characterization, considering that both things are side by side in contemporary formalizations of logic? Now, it might be the case that this sort of view is in fact what Wittgenstein thought he was doing. However, pairing the sort of position McGinn lays out with the general characterization of the meaning of the categories that Wittgenstein himself states, what I have said means in effect that we have no reason to believe a commitment to the explanatory power of syntax would to the meanings of names and propositions as stated. Thus, I do not see a reason to think that we have an alternative explanation here that would problema tize the argument above. I said reasons are for thinking that the mode of signification for variables that represent formal concepts are not explained by reference to categories of ent ities. Thus, if my characterization of Wittgenstein as a thought world metaphysical realist were correct, it would seem that the argument I draw from his commitments would show Wittgenstein as being inconsistent. There may be some truth to this. However,


42 categories of symbols, it may be the case that Wittgenstein is in fact committed to the existence of those categories of entities even if we cannot state those commitments. This would show why it is that we think Wittgenstein is commit ted to ineffable metaphysical truths. It is not enough to say that Wittgenstein does not have commitments to these ineffable truths because Wittgenstein explicitly tells us that we cannot make such g those commitments, one would have to show that what Wittgenstein says about formal concepts gives us reason not to commit him to the prima facie realist argument over and above the fact that Wittgenstein says we cannot make ontological statements. This i s how I am treating formal concepts, but as something that relates to the prima facie realist argument Thus, I am not arguing against McGinn because I am trying to spot argument or an inconsistency in the text. Rather, my reason for so far not being convinced by what she says about rules and syntax is that there is no reason for us to think that this displaces his other commitments about the meani ngs of categories which figure into the premises of the prima facie realist argument § 7 Wittgenstein as a Carnapian There is one other important possibility as to what the emphasis on rules means The thought behind this latter notion might go something like this: there is something significant to focusing on the rules dealings with signs used because the rules are in some vague sense


43 rules for substitution between propositions; w hich is to say that we can amend what propositions count as apart of one class or another and thus what categories of symbols that there are. The divisions between categories of symbols are mutable, and they are mutable because they are in our control. Thu s, there cannot be anything metaphysically interesting about the formal concepts that we have if they are simply a matter of choice. There is evidence that McGinn thinks something along these lines. She says: estion of whether we have given a sign of a certain form an application. It is not a question whose answer is a matter of discovery or experience; it is rather more akin 22 Does Wittgenstein believe that formal concepts are a matter of decision? I believe it would be illuminating to Wittgenstein, eschews metaphysics. This philosopher is Rudolf Carnap I will give my about the topic of thought world metaphysics in TLP. Throug hout his career, Carnap believed that metaphysics could b e avoided. through various phases in which this eschewing of metaphysics was borne out in various ways. The idea that metaphysics can be eschewed because formal concepts are in some imp views as espoused in ESO. It is also worth noting TLP as an influence on his views throughout his career. In his intellectual autobiography from The Philosophy of Ru dolf Carnap, 22 Ibid, p. 188


44 Tractatus Logico Philosophicus. When discussing his adoption of Tarskian semantics in A specific objection which has been raised from the beginning against my approach to semantics is directed against any reference to abstract entities, e.g., classes, properties, numbers, and the like. Some philosophers reject this way of or at least in ne view, which goes back to that of the Vienna Circle and of Wittgenstein, an sentence. 23 Carnap here references the influence that Tractatus Logico Philosophicus had upon him and Vienna Circle. Scholarship on Carnap emphasizes that Carnap was at least earlier metaphysics and nonsense as they appear in TLP. 24 However, this work then states that from the theories of nonsense and semantics in general. 25 Thus, in giving an interpretation and compares it to the views discussed in TLP, it is also worth noting that compare well to aspects of TLP, something that might not be expected given the fact that Carnap is seen to have moved away from Wit In ESO, Carnap argues that debates about the reality of abstract entities such as numbers or propositions can be construed so that commitments to their existence do not need to be construed as metaphysical or onto logical commitments in the sense used by linguistic 23 The Philosophy of Rudolf Carnap, ed. by Paul Arthur Schlipp, p. 65 24 Whether Carnap accurately understood Wittgenstein is a good question. James Conant argues that he did Die berwindung Der Metaphysik Wittgenstein in America, ed. by Timothy McCarthy and Sean C. Stidd. 25 ibid. 24 29. 36) by Andr Carus, in The Cambridge Companion to Carnap for an expositio


45 frameworks and what is internal to those frameworks or external to them. Carnap claims an be construed as either internal to a linguistic framework, which means that it is a claim within an already established framework, or external to a linguistic framework, which means that it is a claim 26 Internal to a 27 Since philosophical questions about the existence of entities are felt to be substantive, Carnap reasons that the philosophi cal question of the existence of numbers is not an internal question but an external question. This external question is non cognitive in nature, and is a question as to the advisability of a framework before it is set up. Carnap believes that the acceptan ce of a linguistic framework is not due to metaphysical argument, but practical decision. In speaking of an external claim as to the reality of the thing world, he says: To accept the thing world means nothing more than to accept a certain form of language in other words, to accept rules for forming statements and for testing, although itself not of a cognitive nature, will nevertheless usually be influenced by theoretical knowled ge, just like any other deliberate decision concerning the simplicity of the use of the thing language may be among the decisive factors. And the questions concerning these qualiti es are indeed of a theoretical nature. But these questions cannot be identified with the question of realism. They are not yes thing language is confirming 28 Carnap then explains what it means to accept a new kind of entity. It involves first 26 ESO, p. 206 27 Ibid, p. 209 28 Ibid, p. 208


46 entities, permitting us to say for any particular entity that it belongs to this kind (e.g., "Red is a property ," "Five is a number "). Second, the introduction of variabl es of the new type. The new entities are values of these variables; the constants (and the closed compound expressions, if any) are substitutable for the variables. With the help of the variables, general sentences concerning the new entities can be formul 29 Thus, a for the type of variables for numbers. Carnap discusses this point in detail in Meaning and Necessity: Suppose somebody constructs a language not only as a subject matter of theoretical investigations but for the purpose of communication. Suppose, further m etc., for which those are substitutable. We see from this decision that he recognizes natural numbers in this sense: and this is the decisive point about numbers in general. He m and and m existence of a prime number. However, the concept of existence here has nothing to do with the ontological concept of existence or reality. The sentence just m between 7 and 13, m same token, we see, furthermore, that the user of the language is willing to recognize the concept Number. Generally speaking, if a language (of ordinary structure) contains certain variables, then we can define in it a designator for the range of values of those variables. In the present case, the m )( m = m m It is important to emphasize the point just made that, once you ad mit certain variables, you are bound to admit the corresponding universal concept. 30 Here we see that Carnap believes the admission of variables implies a commitment to such a universal concept is not any sort of ontological commitment in the typical 29 Ibid, pp. 213 214 30 Ibid, 43 44


47 philosophical sense of a commitment to the existence of a certain ontological category. for the purposes of physics and mathematics involves problems quite differen t from those involved in the choice of a suitable motor for a freight airplane; but, in a sense, both are engineering problems, and I fail to see why metaphysics should enter into the first any 31 I will now spell out in detail t he argument Carnap is making here, or at least what I believe Carnap is committed to given what he has said. First, Carnap thinks we can explicate sentences using numbers, propositions, and the like what I will call as sentences using universal concepts. philosophy Quoting Meaning and Necessity again: The task of making more exact a vague or not quite exact concept used in everyday life or in an earlier stage of scientific or logical development, or rather of replacing it by a newly constructed, more exact concept, belongs among the most important tasks of logical analysis and logical construction. We call this the task of explicati ng, or of giving an explication for, the earlier concept; this earlier concept, or sometimes the term used for it, is called the explicandum; and the new concept, or its term, is called an explicatum of the old one. 32 I propose that Carnap makes traditional ontological categories serve as explicanda for which the corresponding universal concepts serve as explicata. Carnap is not saying that universal concepts are identical to concepts or terms for traditional ontologica l categories, but that we can take sentences involving those categories and give 31 Ibid, 43 32 Ibid, pp. 7 Meaning and Necessity he uses as an explication for the concepts of necessary and analytic truth.


48 explications of them with universal concepts. Further, Carnap believes such universal concepts can be cashed out in appropriate variable styles, as the above quoted passages i ndicate. We can state two points of Carnap as (UC) and (VS) : (UC): Sentences in which traditional ontological categories appear can always be explicated by sentences using universal concepts. (VS): A sentence uses a universal concept if and only if it b elongs to a linguistic framework that employs an appropriate variable style, that is, a kind of variable substitutable for appropriate constants used within that framework. I believe that a consequence of the notion of explication is that an explanation o f a fact involving an explicatum is equally good for a parallel fact involving an explicandum. If a sentence s 1 explicates s 2 the role of s 1 in clarifying the meaning or logical purpose of concepts in s 2 not only removes vagueness or ambiguity in the conc epts involved in s 2 but also serves as a philosophical successor of the concepts in s 2 In general, the role of explication is to create clear and manageable concepts that serve all the same purposes as the concepts we come across in ordinary and scientif ic discourse. The point of explication here is that whatever is a good explanation for something in s 2 is also a good explanation for something in s 1 Turning to our current argument, Carnap believes whatever explanation there is for the universal concepts presented as variable styles in a linguistic framework is equally as good of an explanation for the traditional ontological categories that variable styles explicate. From this fact about explication, along with (UC) and (VS) we get the third step in Car Ontology and Variable Style Equivalence: Ontology and Variable Style Equivalence (OVSE). Any explanation for the fact that some use a language that contains sentences in which traditional ontological categories appear must be an explanation of the fact that they use a linguistic framework that employs an appropriate variable style, and vice versa.


49 for the presence of variable styles, and th us for traditional ontological categories. The metaphysical explanation I have in mind is a thought world metaphysical explanation as explained above. Thus, we can view Carnap as rejecting any realist, irrealist or nominalist thesis that would serve as an explanation of a linguistic framework that employs an appropriate variable style. world metaphysics, we see that underm ine the truth or meaningfulness of the statements. At the beginning of ESO, Carnap claims he is attempting to avoid both Platonism about abstract entities and Nominalism about abstract entities. Platonism and Nominalism correspond to realism and nominalism about the parts of thought that deal with purportedly abstract en tities, respectively. A Platonist accepts the existence of a category of abstract entities as a metaphysical explanation of the relevant features of thought. A Nominalist denies the existenc e of a category of abstract entities as a metaphysical explanation of the relevant features of thought. For Carnap, both Platonism and Nominalism are metaphysical positions one accepting the existence of entities and the other denying it. Carnap grants t hat certain ontological statements are meaningful and can be true, as his view of internal ontological questions reveals. However, I think that his criticism becomes more powerful when is pointed at the attempts of philosophers to use metaphysics to explai n how it is that we have certain concepts within our thought and language in the first place. Getting rid of the desire to attempt to use thought world metaphysics is the importan t way in which Carnap


50 wants to deflate our ontological talk. While Carnap at the beginning of ESO speaks of trying to avoid the truth and falsity of ontological claims such as Platonism and works best if we see him as being critical of the thoug ht world metaphysics I have described above. I also believe we can avoid an irrealist position, since the form of explanation Carnap gives is not metaphysical but what I will We can summarize the above cons iderations as another claim of There is no metaphysical explanation for the fact that some use a linguistic framework that uses appropriate variable styles. Given (OVSE) and the above we can conclude: No Metaphysical Explanation (NME): There is no metaphysical explanation for the fact that some use a language that contains sentences in which traditional ontological categories appear. Carnap also argues that the only explanation for having a linguistic framework with appropriate variable styles is practical decision as to what linguistic frameworks are useful, as can be seen from the quotes from ESO and earlier in Meaning and Necessity. The best explanation for the fact that some use a linguistic framework that uses appropriate variable styles is that it is practically useful for those people to use that framework. Given (OVSE) and the above we can conclude: Pragmatic Explanation (PE): The fact that some use a language in which traditional ontological categories appear is explained by the practical utility of a linguistic framework that uses appropriate variable styles in order to explicate that language. It is important to note that (PE) and (NME) are mutually supporting. If (NME) is true, then whatever explanation there is for traditional ontological categories is not


51 metaphysical, and (PE) is that sort of explanation. If (PE) is true, we do not have a meta physical explanation for traditional ontological categories. Thus, (NME) follows. In ESO, it seems as if Carnap uses (PE) to defend (NME) Either way, the positions are mutually supporting, and the argument ends the same way: the explanation for traditiona l ontological categories is not metaphysical, but is the practical choice of the engineer of a linguistic framework that explicates those categories. As stated above, b ecause the explanation given is a pragmatic one, we do not have any sense in which Carna position is a thought world metap hysical thesis whether realist, irrealist or nominalist. It is worth noting that Wittgenstein compares well to Carnap on some of the above theses. Barring possible differences between the two philosophers, I believe we can substitute (UC) However, what is to be substituted for clarification of thoughts. Philosophy is not a body of doctrine but an activity. A philosophical work consists essentially of elucidations. Philosophy does not result in clarification comes by way of a conce ptual notation that puts language into a notation that


52 different symbols and by not using in a superficially similar way signs that have different modes of signification: tha t is to say, a sign language that is governed by logical grammar by signifying clearly what cannot be said by stating clearly what can be said (4.115). When we have done so, w e recognize that much of what passes as philosophy is nonsensical, as it cannot be presented as dictated above. When Wittgenstein says that formal concepts cannot be expressed in propositions with sense, he means that any philosophical statements that do u se traditional ontological concepts come out as formal concepts in the conceptual notation that Wittgenstein advances. attempt to replace ordinary notions with different yet related ones. After all, Wittgenstein does. However, the relevant point of comparison is that both methods are ways of casting language into logical notation that attempt to get us to a proper understanding of the workings of our language and how certain concepts work in it. For Carnap, explicating sentences involving traditional ontological categories brings us to an understanding of how it is that they are merely a reflection of our practical decision. For Wittgenstein, elucidating sentences involving traditional ontological categories brings us to an understanding of how it is that such make manifest the form of language and that some such statements involving them are nonsense


53 With these two points developed, I argue Wittgenstein believes something parallel (UC) ; namely, (FC) : (FC): Sentences in which traditional o ntological categories appear can always be elucidated by sentences formal concepts. sopher explicitly claims that all traditional ontological categories can be dealt with in the way I have argued that each of them does for concept frameworks at will, the amount of categories to explicate does not have a hard and fast limit. However, Wittgenstein in TLP often speaks of language, and all of language, admitting of one specific sort of conceptual notation. Because of this, the variable styles that we come to are not flexible in number. Thus, it will come out that some concepts that are or resemble traditional ontological ones will appear in properly notated language, and that statements involving them come out to be nonsense. This becomes a problem when we realize th at the beginning of TLP seems to make claims that the passages in 4.126 criticize basically the entirety of metaphysics because of the liberality of explication, Wittg enstein seems to only be able to criticize the metaphysics he himself engages in. Nonetheless, we can provide some evidence Wittgenstein had in mind a larger class of ontological talk. For one, there are philosophers who speak of the existence of roughly the sorts of things Wittgenstein is talking about. A good example of such a philosopher is Bertrand Russell. In The Philosophy of Logical Atomism Russell declares


54 that there are facts 33 and particulars 34 much the same way Wittgenstein says that it is nonse nse to speak of objects and facts. Also, if Wittgenstein is correct that his conceptual notation is the key to understanding all of language, then all of the language of the metaphysicians must come out to have some such formal concepts in them. Although t his would make metaphysicians who do not talk about particulars and facts come out to be saying things quite different once what they say is placed in conceptual notation, it makes t metaphysicians at best are saying are the categories of the world, and then showing how metaphysicians say nonsense at their best. Whatever one thinks of this, all of this is to say that the scope of the traditional ontological categories in (FC) is broa d enough to make a (UC) (VS) important to emphasize the point just made that, once you admit certain variables, you are bound to admit the corresponding uni object falling under it constants of names shows the existence of the pseudo way that Carnap claims having variables substitutable for constants of numbers commits one t o the exi stence of the concept Number. There are certain differences, of course. For example, Wittgenstein does not use lambda abstraction like Carnap to capture the idea of 33 Lecture 1, esp. pp. 40 47 34 Lecture 2, esp. pp. 54 63


55 However, I believe they are sufficiently similar on this point. So, I think we can state (VS*) (VS*) A sentence uses a formal concept if and only if it employs a style of variables in language placed in conceptual no tation, that is, a kind of variable substitutable for appropriate constants. There is an important difference between Carnap and Wittgenstein that may make it seem as if a fruitful comparison between the two fails. Carnap thinks that traditional ontologic al claims are internally analytic and trivial, and externally non cognitive. clai ontological claims are nonsensical in the way that Carnap thinks they are non cognitive. Wittgenstein thinks they are nonsensical because they attempt to say what can only be shown in the logical form of language. Carnap, on the other hand, thinks they are non cognitive because an external claim can only sensibly be worded as a practical recommendation as to the reason to adopt a linguistic framework. For Carnap, the distin could be dropped in any description of reality, as Wittgenstein believes that propositions that describe reality are all analyzable into concatenation of names, and names are the internal ontological statements are trivially true and analytic. For one, nonsense cannot be


56 true. More importantly, Wittgenstein has a separate term for trivial, analytic statements: senseless ( sinnlos ). These are the tautologies and contradictions of logic, as well as the they contain mea ningful constituents. However, tautologies and contradictions cannot say anything meaningful because, according to Wittgenstein, a meaningful proposition must tautology such as ( p or ~p ) cannot possibly be false, and thus cannot convey a sense. 35 In short, Wittgenstein and Carnap differ on how they view the semantics of the statements of traditional ontology they are trying to understand. Given that Wittgenstein declares the metaphysical portions of his book to be nonsense, we might be led to think that the key to understanding what Wittgenstein really is talking about in his more explicitly metaphysical moments of TLP is to understand what he thinks of nonsense. Thus, it would seem quite unfruitful to compare Wittgenstein to the later Carnap, a philosopher who seems to differ on how he views metaphysics to be nonsense. However, I believe the value in the comparison comes in trying to determine, given that Wittgenstein beli eves (FC) and (VS*), whether Wittgenstein believes something that is (NME) This, of course, is just what I have been doing so far. The reason this comparison is important at this point of the investigation is that McGinn seems to be giving (PE) and similarly is using something like (PE) (NME) Of course, I am 35 For an example of a discussion on the distinction between sinnlos propositions and unsinning propositions see Chapter 1 of Ian P


57 fleshed out, well known way of making this brief point, the comparison is useful in order to see how the prima facie realist argument relates to it and to see whether this more fully fleshed out view is a plausible interpretation of Wittgenstein which woul d challenge the interpretation given by that argument. The challenge thi s presents to the prima facie realist argument is not at any of the po ints so far given. Namely, Carnap would (3) (5 ). If the explanatio n for the fact that there are different variable styles is the practical utility of a linguistic framework, there is no need to explain the difference between variable styles by an ontological commitment to those categories of entities, even if we might th ink that we can be committed to these entities in the trivial, internal sense. If Wittgenstein is a Carnapian in this respect, how do we relate the points drawn from ESO above to what has been said about TLP? For one, there would be a slightly different version of (PE). we get (PE*): (PE*) The fact that some use a language in which traditional ontological categories appear is explaine d by the practical utility of the conceptual script that uses appropriate vari able styles in or der to elucidate that language. As mentioned before, Wittgenstein does not allow explicit commitments to categories of entities as Carnap does, meaning that the two are not comparable in at least this respect. There is also an area in which they are not comparable at this point i t is unclear whether the way in which parts of


58 language are meaningful 36 However the point of the comp arison here is that even if explanation by w ay of categories of entities is an explanation at least in a limited sense of some sort for the meaningfulness of particular signs and for the categories of symbols the explanation that fits the best is (PE*). If (PE*) is a commitment to be found in Wittgenstein, it is something the text will bear out. However, it is important here to be clearer on what extent (PE*) supports (NME). (PE) and (NME) are mutually supporting. To be su re, there may or may not be some argument there as to why they support each other, but my goal above was merely to demonstrate s. As said above, the Carnapian position hedges on the idea that a commitment to a category of entities due to the content of categories of thought in any way is a metaphysical explanation. Implicit here is the idea that any commitment to those entities is in some way uninteresting or trivial. This is the line that Carnap takes, claiming that internal ontological statements are trivial, analytic truths. the prima facie realist argument given above. If Wittgenstein is committed to (PE*), and (PE*) supports (NME), it would gi ve us rea son to deny that Wittgenstein is committed to (5) in the prima facie realist argument, as (5) is a metaphysical explanation. Of course, it is possible Wittgenstein is committed to (PE* ) and (5) and thus has an inconsistent view. 36 When laying out the basics of his semantic theory at the beginning of Meaning and Necessity, Carnap entity is frequently used in this book. I am aware of the metaphysical connotations associated with it, but I hope that the reader will be able to leave them aside and to take the word in the simple sense which it is meant here, as a common designation for properties, propositions, and other intensions, on the one hand, and for classes, i does not think that his use of the term entity commits him to a metaphysical view, a view presumably like the one given in the prima facie realist argument. Thus, we can take it that his reasons for thinking he has ESO, and thus are the reasons I have been discussing.


59 However, considering that the prima facie realist argument drawn is a prima facie one, and considering the ways in which Wittgenstein has been interpreted by many readers as viewing one thing or the other due to the relative obscurity of the work, it is better for the purposes of t his thesis to draw out a view out of Wittgenstein that is consistent as possible rather than worry if the commitments we are drawing o ut conflict. As I said in Section 2 the commitments I have drawn out are attempts to answer our questions and views about metaphysics that we see in the book given our own questions, setting aside the issue of nonsense in the book. As such, I do not believe it is important here to consider if those commitments conflict, given that those commitments are drawn out for the expr essed purposes of finding consistent views and arguments that we can discover and assess given our philosophical preoccupations. What is implicit in (PE*) (and (PE) ) is an idea I will call Pragmatic Pluralism. Given that the explanation for having traditi onal ontological concepts is the practical utility of a chosen framework, this means that there exists the possibility of the choice of different and equall y adequate conceptual scripts with different formal concepts corresponding to those categories Conn ecting this to how the prima facie realist argument is phrased, there can be languages that have or lack certain categories of symbol s that other language have. In terms of the prima facie realist argumen t, there can be a language that has A and another language that does not have A. Pragmatic Pluralism is the following : Pragmatic Pluralism: a) It is possible for there to be languages L1, L2, and L3, such that L1 contains categories of symbols A and B, L2 contains A and does not contain B, and L3 does not contain A and contains B, and L1, L2, and L3 are all adequate languages. b) T he fact that L1, L2, and L3 contain the categories that they do is explained by the preferential choice of the engineer of that language.


60 meets certain standards for an adequate linguistic conceptual scripts that are fully elucidated. Recall (5): (5): Thus, the explanation for the fact that members of a category of symbols A differ from members of a category of symbols B is that the members of A are about some category of entities A* different from the category of entities B* that the members of B are about. Pragmatic Pluralism implies (5) is false i f th e following is true: the fact that the presence of categories of signs in possible adequate languages L1, L2, and L3 is a matter categories of signs does not come f rom the entities that they are about. The pragmatic pluralist can affirm that the categories are about something in some sense. However, the sense given in (5) is stronger: the entities that the categories of thought are about explain in an importantly metaphysical wa y why the categories differ in meaning. The intuitive idea here, the one that Carnap had, was that there is no reason to commi t oneself to fundamental ontological entities to explain the differences in me aning of categories of signs if those signs are present in languages in the first place by the non cognitive, preferential choice of a language engineer. One could wonder if this intuition is sound. It may not be true that just because some categories of symbols are present in a language due to preferential choice that the meaning of those symbols cannot be explained by the existence of categories of entities. However, it is at the very least plausible. It seems that if we think there is an explanation for the categories of symbols in a language in terms of categories of entities, a


61 commitment to those entities as metaphysical explanantia makes those entities responsible for the presence of symbols in a language if the entities are responsible for the meani ng of those symbols when present. Whatever one thinks about this, what I have attempted here is to show how the views of another influential philosopher like Carnap sheds light on the claim implicit in McGinn that the rules for formal concepts are somehow that there is no metaphysical explanation for the presence of formal concepts, and that they are rather up to preferential choice. § 8 Is Wittgenstein Really a Carnapian? Now the question becomes the following: Does Wittgenstein ascribe to Pragmatic Pluralism ? There are parts in TLP where Wittgenstein states that there is some sort of preferential choice in the logical structure of language. This comes out in his discussion of scientific explanation. a priori belief in a law of conservation, but rather a priori knowledge of the possibility of a logical form. All such propo sitions, including the principle of sufficient reason, the laws of continuity in nature and of least effort in nature, etc. etc. all these are a priori 6.34). He goes on to compare Newtonian mechanics to a mesh imposed upon a surface of white surface with an irregular placement of black dots. Each hole in the mesh is of a certain shape and size. If the holes in a mesh are one square inch squares, then each part of the surface that is demarcated by one square inch of a square mesh will be either black or white. The analogy at work is that Newtonian mechani cs is one of many different mesh nets


62 (6.341) does tell us something about it is the precise way in which it is possible to describe it by The implica tion here is that a certain kind of science uses sentences that have distinctive logical form. The use of the metaphor of the mesh and his saying that it does r of logic being (4.023). Because of the connection between logical form and formal concepts, it is not a large jump to conclude that science may have its own arsenal of categories used to describe the world. have achieved the same result by using a net with a triangular (6.341). All of the above it is optional as to what formal p r operties a language may have. However, even if we admit the possibility of different categories of symbols here, it seems that there is no variability in names and propositions, which just are the categories that garner the commitments in question. Th ere are scattered passages that seem to supp ort this conclusion. Take 6.124 as an example: d elementary propositions sense; The implication is that this holds for all propositions of logic (all tautologies and contradictions), not some here and some there. The more important point is that the general tenor of the book leads us to believe that


63 function of elementary we do not seem to have any variability in choosing different conceptual scripts that could include or exclude names or pr opositions. M cGinn agrees with this assessment: some concept of elementary propositions quite apart from their particular logical TLP 5.555). The concept of element ary propositions that we have constitutes everything that is essential to the notion of a logical picture, that is, everything that belongs to the nature of a proposition that represen ts a possible state of affairs. 37 McGinn hinted before that there the ch oice between formal concepts could be a matter of decision. However, she here agrees with my assessment that elementary propositions and names are not something that could be a matter of decision and are essential to the representative aspect o f language. The way in which I differ from McGinn is that she does not believe, as her introduction shows, that Wittgenstein has any metaphysical commitments that follow from this essence of representation. The reason she holds this belief is because it co a priori the forms of elementary propositions. Right after the quote where she says that there are un analyzable subject predicate propositions, as if this were a matter of predicate propositions depends only on whether we have introduced signs of a certain 37 Elucidating the Tractatus, 188.


64 form in proposit i 38 Here she is implicitly referring to s claim that we cannot know a priori the forms of elementary propo sitions although we at least know that there are elementary propositions with names in E lementary propositions consist of names. Since, however, we are unable to give the number of names with different meanings, we are also unable to a reason to believe that there is a difference between the explanation for the presence of formal concepts and the explanation for the presence of elementary propositions and names. Formal concepts are akin to a matter of decision, whereas elementary propo sitions and names are essential to the representative nature of language. However, the problem is that the re are two formal concepts that correspond to elementary propositions and names: objects and states of affairs 39 If elementary propositions and names ar e not a matter of decision, then the formal concepts to which they correspond are also not a matter of decision. The formal as a formal concept is present in a langua ge insofar as there is a category of symbols with the same formal properties. Elementary propositions and names are just those categories. Therefore, at least those two formal concepts are not a matter of decision. Thus, McGinn cannot consistently hold tha t formal concepts are a matter of decision and then claim that elementary propositions and names are not. 38 Ibid. 39


65 § 9 A Final Challenge to the Prima Facie Realist Argument At this point, I believe the prima facie realist argument has held against the kinds of obj ections found in McGinn. I believe that since these objections cover a wide range of possible ways of reading TLP realism, we may have a higher degree of confidence that Wittgenstein is committed to a sort of realism in TLP. However, there is an important consideration I will consider that would work against such a read ing. In short, I think that although we do not have reason to deny that there is thought world metaphysics at work in TLP for the sorts of reasons given by McGinn that I have discussed above i t is unclear whether it needs to be a realist thesis or an irre alist one. As such, if there is no way to decide between a realist or irrealist thesis as the best interpretation of the thought world metaphysics in TLP, it remains unclear if we should say that Wittgenstein is committed to thought world metaphysics. The prima facie realism argument incurs a commitment to the existence of categories of entities because these entities expla in the meaningfulness of thought. However, we could word the steps of the prima facie realist argument as a prima facie irrealist argum ent. How could we make the conclusion an irrealist one? (1) and (2) would remain the same in this irrealist argument. Call them (1I) and (2I) : (1I) A symbol x is different from another symbol y if and only if x differs from y in mode of signification, or (2I) Members of category of symbols A are different from members of category of symbols B if and only if the members A differ from the members of B in However, (3) becomes (3I): (3I): T here being entities which symbols are about is explained by those symbols having mean ing


66 Then, we get (4I): (4I): Thus, the explanation for the fact that there is an entity a and a different entity b is the fact that there are symbols x and y which are about a and b, respectively. [From 1, 3I] Finally, we get (5I): (5I) The explanation for the fact that there are members of a category of entities A* and members of a different category of entities B* is that the members of a category of signs A are about some category of entities A* different from the category of entities B* that the members of B are about. [From 2, 4I] This argument does not seem to intuitively match up with the way we would read es and propositions. We might side with the realist interpr etation because it seems to fit better with how we intuitively think the direction of explanation would go. We are much more likely to think that entities in the world do the explanation of the mea ning rather than the meaning doing the explaining of the existence of those entities. The only way we could secure such an interpretation is if we had some evidence that Wittgenstein claims that the explanatory direction goes from world to thought rather t han the other way around. However, I have culled theses dealing with explanation from propositions of TLP that merely state how it is that certain segments of language mean what they do. My inser realist argument is ju stified only by the thoughts above This is why the prima facie realist argument is prima facie objects and that the sense of propositions is the existence and non existence of states of affairs, we are na turally led to think that the world is the side of the coin doing the work. Yet, it does not have to be that way.


67 What would we have to get out of TLP such that it would have to be that way? This requires getting clearer on what it is for a metaphysical explanation to go one way or another. In other words, it would have to be a way of clearing up the vague metaphor that on thought rathe r than its being Such a way is suggested in how I have described why a thought world metaphysical explanation is a metaphysical explanation. For t he realist, certain features of the categories of t hought are not possible without the existence of entities corresponding to the categories of thought. For the irrealist the existence of entities corresponding to the categories of thought is not possible without the categories of thought with certain fea tures. The best way I see in respect to these theses is to note that, for the realist, the existence of categories of entities is possible without the existence of the ca tegories of thought. It seems that a realist would think that the existence of the categories of entities does not require our having thoughts which have such corresponding categories. The irrealist by contrast, is committed to there being such categories of entities only if there are categories of thought corresponding to them. The irrealist need not think that nothing exists outside of thought, broadly speaking, but that such things do not come into the categories they do without their being correspondin g categories of thought. What in TLP would show Wittgenstein believing that such categories of entities could exist without there being categories of thought? It may seem that we could secure a realist commitment ositions. Take his early


68 possible We here have a characterization of objects one of the categori es of entities in question that does seem to give it qualities that extend to all possible worlds. We also have propositions about states of affairs and about how they relate to states of affairs. In addition to 2.0124, there What is the case a fact, is the Thus, objects, which are necessary existents, combine to form states of affairs. A state or states of affairs in the actual worl d is a fact, a state or states of affairs not in the actual world is a combination of objects in a possible yet not actual world. It seems then that names, which mean their object, and propositions, whose sense is a representation of a ing or not being in the actual world are ways in which we as humans in the worlds which we inhabit connect to entities which exist in the actual world, possible worlds we inhabit, and then possible worlds which we do not inhabit. However, Witt genstein say s other things that imply that the characterization of objects and states of affairs is bound to the way our language works. This seems to be the p 5.64: The limits of my language mean the limits of my world. Log ic pervades the world: the limits of the world are also its limits. would appear to presuppose that we were excluding certain possibilities, and this cannot be the ca se, since it would require that logic should go beyond the limits of the world; for only in that way could it view those limits from the other side as well. (5.6 5.61).


69 For what the solipsist means is quite correct; only it cannot be said but makes itself manifest. (5.62). Here it can be seen that solipsism, when its implications are followed out strictly, coincides with pure realism. The self of solipsism shrinks to a point without extension, and there remains the reality co ordinated with it. (5.64) These are difficult passage s and they are passages that are perhaps taken out of context. Considering the diversity of interpretations of TLP in general, any reading of these passages would probably be as controversial as the status of t he metaphysical propositions in 1 and 2 seem to be. It is unclear in the context of the present discussion metaphysical propositions in 1 and 2. However, it seems at least a good en ou gh reason to doubt that we are obligated to interpre t Wittgenstein as siding as either a realis t or irrealist We may read such a passage as being reason to believe that Wittgenstein does not believe categories of entities exist without their being such categories of thought, and thus read Wittgenstein as a thought world metaphysical irrealist However, my claim here is that we have as much weight to believe this as we do for believing that such categories of entities exist outside of thought, something t hat seems to follow from simply reading propositions 1 and 2. Making a case either way would require making a case on thin interpretative grounds given how easily one could characterize the direction of explanation as going either in a realist or irrealist direction. Perhaps one might emphasize the fact that the passage above about solipsism is later in TLP and thus should make us tions 1 and 2 that go before it, but this is tendentious. As opposed to bringing some support for an irrealist reading, I think 5.6 nicely encapsulates what it would mean to read Wittgenstein as a realist or an irrealist and how


70 5.6 itself might put The limits of my language mean the limits of my world. Logic pervades the world: the limits of irrealist that the li mits of my language entail the limits of my world, and not the other way my However, the fact that it could be read in both of these ways, and the fact that coincides explanation is to be found, and that perhaps it is intended that Wittgenstein does not think that any such direction of explanation will work. The fact that there is no direction of explanation given means that Wittgenstein cannot sensibly be called a realist or irrealist This, then, is the way in which we would hedging on the use of If there is no way to decide which way the direction of explanation goes, we cannot say that Wittgenstein accepts the existence of a category of entities in order to explain the features of thought if it is also possible to say that the features of thought could explain the accepted existence of a category of entities. Could Wittgenstein then be a nominalist? Does Wittgenstein deny the existence of a category of entities that would either explain features of thought or be explained by features of thought? This does not seem to be the case either. Nothing in the text gives us


71 the idea that there are no objects or states of affairs. This should be evident from the previously quoted passages. If Wittgenstein is not a realist, irrealist or nominalist, then Wittgenstein is not committed to any thought world metaphysics. Clearly, however, he is not a thought world metaphysician in a sense different from the sense implied by McGinn or the sense in which Carnap was not a thought world metaphysician. I have argued that we have no reason to read Wittgenstein as explaining the presence of formal concepts by way of the categories of symbols is explained b cannot do any metaphysical work as I have described thou ght w orld metaphysics, given that the entities are not posited in order to give some sort of direction of explanation. However, this conclusion may still seem odd. Although I have rejected the notion that Wittgenstein is committed to some kind of metaphysi cs, we are still left wondering why Wittgenstein talks about objects and states of affairs at all. They seem to have something to do with the meaning of categories of symbols, even if what they have to do with the meaning of categories of symbols is not me taphysical in the sense that I have explained. There seems to be a sense of metaphysics that Wittgenstein does still commit himself to, although it is a weaker sense at that. Perhaps we could describe the weaker sense in the following way. In TLP, t here st ill seems to be a parallelism between what exists and what the features of thought are. Yet, this parallelism cannot be explained as thought world metaphysics. The parallelism we see would make us think that Wittgenstein has some sort of metaphysics whereb y the parallelism is explained by the entities or where the features in thought that are paralleled


72 explains the existence of those entities. However, if we cannot determine the way in which the direction of explanation goes, all we have is a certain sort of parallelism. This parallelism makes us think that certain entities exist, although no claims are made on why this is the case. an edited version of the prima facie realist argument: (1 ) A symbol x is different from another symbol y if and only if x differs from y (2 ) Members of category of symbols A are different from members of category of symbols B if and only if the members A differ from the members of B in (3 ) A symbol means what it does if and only if that symbol is about something (4 ) Thus, a symbol x differs from another symbol y if and only if x is about someth ing different than y. [From 1, 3] (5 ) Thus, members of a category of symbols A differ from members of a category of symbols B i f and only if the members of A are about some category of entities A* different from the category of entities B* that the member s of B are about. [From 2, 4] The argument was originally stated as saying that the meaning of symbols and categories of symbols is explained by the existence of entities and categories of entities. This explanation was cashed out as thought world metaphy sical realism. However, given the fact that we cannot rest on this notion of explanation if it could also be cashed as irrealism Instead, categories of symbols are different from each other just if and only if they are it is j ust a fact about what it is for names and proposi tions to mean what they do. That fact somehow does the required explanation. Thus, metaphysics in TLP is metaphysics in a


73 weak er sense because we are just saying that there are certain entities that correspond to thought while not making any other sorts of claims. Now, propositions 1 and 2 of TLP seem to be other sorts of claims about those entities. The propositions contain des criptions of objects and their properties and how those properties interact in produ cing states of affairs. I believe that it is correct to note world and logic to share the same multiplicity encapsulates this idea. However, what I have been arguing is that we do not have a sense in which the world or logic explains why there is such a parallelism, and thus we cannot say whether the metaphysics of propositions 1 and 2 explains the logic of the later propositions or the other way around. Conclusion This paper started by stating that Wittgenstein seemed to be committed to a certain kind of metaphysics. After considering and rejecting interpretations that do not find h im as having this commitment, I have ultimately concluded that he does not have such a commitment. This may lead us to wonder what to think about metaphysics in TLP in general. Early on I argued that the possibility of there being thought world metaphysics in TLP is central to how the metaphysics of the rest of the book works. The idea that the world impresses its features on thought would explain why it is that Wittgenstein felt the need to discuss the nature of the world in Propositions 1 and 2 in the fir st place. If we do not find thought world metaphysics in the book, it may make us wonder what the point of those passages really is. Otherwise, their presence in the text may seem mysterious.


74 I can only here offer a suggestion. Describing how the world is, stating that thought represents reality insofar as it is a picture of reality, and then describing the logical structure of thought is a natural way of motivating what seems to be Wittgenstei to set the limits of thought. If we first characterize in general what thought is about, and then see how what thought is about parallels the structure of thought, we are in a better position to see how logic must wor k if it is limited to the reality described at the beginning of TLP. Of course, I am ignoring the fact that Wittgenstein finds the entire text to be nonsense. This fact seems to be the elephant in the room of my entire paper. However, if we set aside the idea that Wittgenstein is saying plain nonsense as I argued that we ought to do, we set aside the idea that Wittgenstein is being deceptive in his use of metaphysical statements Thus, there is a point to the metaphysical statements of the text at the begi nning. I have offered my one suggestion. Also, as I stated at the end of Section 9, we do get the idea that even though we do not have thought world metaphysics, we can still TLP sets the limi ts to thought, we can give characterizations (albeit nonsensical ones) of what thought is about, even though there is nothing over and above our describing what thought is about in terms of some metaphysical explanation about how the relationship between t hought and reality works. Thus, even given my suggestion and my interpretation of a weaker form of metaphysics in TLP, there is still a degree of mystery to why it is that thought and the world are related in the way that they are. I have argued that Witt genstein does not give an answer as to why they are related This is mysterious because we might think that this


75 is something that cries out for explanation. Thus, perhaps the value of my interpretation of TLP could come in thinking about whether this lack of explanation is a good or bad thing. At that point, one would also be thinking about the relationship between thought and reality beyond just interpreting TLP. Surely, it is a positive for a work in the history of philosophy that requires one to think m would suggest that my paper is beneficial for this reason. I hope that the notions I have introduced and the interpretations and comparisons I have made are starting points for further reflections on the issues of philosophy.


7 6 Bibliography Carnap, Rudolf. Meaning and Necessity. Chicago: University of Chicago Press, 1964. Diamond, Cora. The Realistic Spirit. Cambridge, MA: MIT Press, 1991. Friedman, Michael and Creath, Richard, eds. The Cambridge Companion to Carnap. Cambridge: Cambridge University Press, 2008. Hobbes, Thomas. Leviathan. New York: Barnes and Noble, Inc., 2004. Originally published in 1651. Johnston, Colin. Synthese (2009) : 167: 145 161 Klbel, Max and Weiss, Bernhard, eds. The Lasting Significance of Wittgenstein's Philosophy. London: Routledge, 2004. McCarthy, Timothy and Stidd, Sean C., eds. Wittgenstein in America. Oxford: Oxford University Press, 2001. McGinn, Marie. Elucidating the Tractatus. Oxford: Oxford University Press, 2006. Proops, Ian. New York: Garland Publishing, Inc., 2000. Putnam, Hilary. Proceedings and Ad dresses of the American Philosophical Association, Vol. 50, No. 6 (Aug. 1977), pp. 483 498 Putnam, Hilary. Reason, Truth, and History. Oxford: Cambridge University Press, 1981. Russell, Bertrand. The Philosophy of Logical Atomism. Peru, Illinois: Open C ourt Publishing, 1985. Schlipp, Paul, eds. The Philosophy of Rudolf Carnap. Carbondale, Illinois: The Library of Living Philosophers, Inc., 1963. Wittgenstein, Ludwig. Tractatus Logico Philosophicus. Translated by Pears, David and McGuinness, B.F. London: Routledge and Kegan Paul, Ltd: 1963.

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EYI71FFHL_05NFIN INGEST_TIME 2017-06-01T19:02:32Z PACKAGE AA00057805_00001