The Paradoxical Reading of The Tractatus

(1)

The Paradoxical Reading of The Tractatus

MSc Thesis (Afstudeerscriptie) written by

Fernando Ortega Ruiz

(born 28th February 1987 in Seville, Spain)

under the supervision of Martin Stokhof , and submitted to the Board of Examiners in partial fulfillment of the requirements for the degree of

MSc in Logic

at the Universiteit van Amsterdam.

Date of the public defense: Members of the Thesis Committee: 29th September 2016 Prof Dr Benedikt Loewe

Prof Dr Martin Stokhof Dr Tamara Dobler Dr Luca Incurvati

(2)

Abstract

The aim of this paper is to provide an interpretation of Wittgen-stein’s Tractatus Logico-Philosophicus based on the idea that it is a paradoxical work, and it can only be analyzed from a perspective that assumes this paradoxical character and does not attempt to avoid it. The debate on nonsense will be used as a starting point, insofar as it treats in detail the paradox, but the interpretation presented in this paper will oppose the two readings that result from the debate, as both of them seek a way out of the paradox.

This interpretation will be grounded on a transcendental notion of limit, and on the idea that the limit shows itself in the ‘act of represen-tation’. The thesis of this paper is that, following this interpretation and the elucidatory notion of philosophy presented in the book, it is possible to read the Tractatus Logico-Philosophicus as a paradoxical, and yet significant work.

Acknowledgements

Firstly, I would like to express my sincere gratitude to my supervisor Prof Dr Martin Stokhof for his patience, motivation, and guidance. He renewed my love for the philosophy of Mr. Wittgenstein, and I could not be more grateful for this.

Besides my supervisor, I would like to thank the rest of my thesis committee: Prof Dr Benedikt Loewe, Dr Tamara Dobler, and Dr Luca Incurvati, for taking the time to read my thesis.

I thank Nigel Sequeira, Kostis Gickas, Pablo Oldaq, Dan Frumin, Kuan Ko-Hun, Ana Lucia Vargas, Aldo Ramirez, Moritz Baeumel, Rick Bellaar, Sarah Hiller, Pietro Pasotti, and the rest of my fellow logic students, for their help and friendship. I also want to thank my flatmates, Coco and Mona, for being there in the good and the bad times, and Paolo Galeazzi and Hugo Nobrega for the amazing football matches we played together.

Last but not the least, I would like to thank my family and my friends from Spain: their support throughout these years has been in-valuable.

(3)

Introduction

The Tractatus Logico-Philosophicus (henceforth TLP),1by Ludwig Wittgen-stein, is a pivotal work in the philosophical landscape of the twentieth cen-tury. Moreover, it has been significant in several fields, such as logic, se-mantics, cognition, and ethics. Commonly regarded as a piece of analytic philosophy intertwined with elements from continental philosophy, TLP is undoubtedly an overly complex work. It has motivated many debates and, to date, there is not general consensus as to how it should be read.

One of the most relevant aspects of TLP is its unquestioned paradoxical character. It develops a theory of sense which concludes that the theory itself is nonsensical. This particular feature of TLP has produced one of the most influential and extended debates, the debate on nonsense, which has become a crucial starting point for most interpretations of the book nowadays – either defending one of the two sides of the debate or opposing both of them.

Our goal in this paper is to present an interpretation of TLP which takes the latter position – i.e., an interpretation that confronts the two sides of the debate. The reason is the following: we believe that the two positions have something in common, namely that both of them try to avoid the paradoxical character of TLP in order to provide a non-contradictory interpretation of the book. However, we believe that the paradox is essential to the theses presented in TLP, and that it cannot be properly understood if it is not from the ground of the paradox itself. Hence, we will refute the two interpretations of the debate; we will show that, actually, they end up offering partial and inconsistent readings of the book; and we will provide our own interpretation of TLP, explaining why the paradox is essential for the understanding of TLP.

The structure of the paper will be the following:

In Chapter 1 we present a reading of TLP at face value, trying to avoid as far as possible interpretations beyond what is actually written in TLP. We focus on four main topics: the picture theory of meaning, the notions of form and structure, the distinction between ‘what can be said’ and ‘what shows itself’, and the role of philosophy in TLP.

In Chapter 2, we present the debate on nonsense. We provide an expo-sition of the main ideas that characterize each poexpo-sition – the metaphysical

1

We will use the translation by C. K. Ogden (Cf. Wittgenstein (1922) for reference on the edition). There are some terms which translation is problematic in this edition. For this terms, we will use our own translation (in particular, cf. note 33).

(5)

reading,2 and the therapeutic reading.3

In Chapter 3, we evaluate these two readings.4 We defend our view – that they represent two different attempts to solve the paradox – and we show why they are not suited to offer a proper interpretation of TLP.

Finally, in Chapter 4 we present our own reading, which we call the paradoxical reading of TLP. First, we present our interpretation of the crucial ideas of TLP, through the notions of act of representation and transcendental limit. Then, we explain how to analyze the paradox from the ground of our own interpretation, and what are the advantages of the paradoxical reading of TLP.

2_{The main references for this interpretation are Anscombe (1959), and Hacker (2000).}

Other relevant sources are: Winblad (1993), Lippit and Hutto (1998), Kremer (2001), Vilhauer (2003), and Hutto (2006).

3

The main reference for this interpretation of TLP is Crary and Read (2000). Other relevant sources are: Diamond (1988), Read (2005), Conant and Diamond (2004), Read and Deans (2003), Hutchinson and Read (2006), and Conant (2007).

4

As we mentioned, we take a critical position regarding the debate, opposing both the metaphysical reading and the therapeutic reading. Other relevant analysis of the debate on nonsense are: Reid (1998), McGinn (1999), Proops (2001), Emiliani (2003), Brand (2004), Morris and Dodd (2007), and Stokhof (2011).

(6)

1 TLP at Face Value

The goal of this Chapter is to introduce the main ideas presented in TLP, as a means to prepare the ground for the debate on nonsense. The debate, which we will treat in Chapters 2 and 3, is meant to provide us with a perspective on the paradoxical aspect of TLP, indispensable for our own interpretation. Our task in this Chapter is to provide a brief explanation of the main ideas TLP, in relation to the notion of nonsense. The aim is to outline a reading of TLP at face value. Of course, we are aware that any reading is already an interpretation, but the idea is to remain as close as possible to what is actually written.

Since our goal is to set the ground for the debate, our summary will focus on the ideas that we consider most relevant for it. Hence, the reading on this Chapter should not be considered an exhaustive analysis: we will leave out some important issues, such as the mystical aspects presented in TLP or its ethical views.5 _{We will try to avoid taking sides in the debate,}

for the moment. However, as we will see, different readings put more weight in some particular statements of TLP and disregard others. For this reason, our own reading of some particular ideas of TLP might be closer to one of the two positions in the debate. This will be clear in the more extensive analyses in further Sections.6

We will divide our overview of TLP in four main themes: the picture theory, the notions of form and structure,7 the distinction between saying and showing, and the status of TLP as a (meta)philosophical work.8 Before getting to these topics, we must say a word about the aim of Wittgenstein in writing TLP. In the Preface, Wittgenstein writes the following:

The book deals with the problems of philosophy and shows, as I believe, that the method of formulating these problems rests on the misunderstanding of the logic of our language. Its whole meaning could be summed up

some-5

We will briefly treat the notion of ‘the mystical’ in Sections 1.3 and 4.1. We will not treat the ethics of TLP. The reason is that it is a topic which requires an extensive analysis, and hence, it goes beyond the scope of this paper. Note that the ethics of TLP is a topic which analysis might be affected by the paradoxical interpretation of TLP; however, it is not particularly relevant for the analysis of the paradox in turn. For a detailed examination of the ethical aspect of TLP cf. Stokhof (2002), Arnswald (2009), and Iczkovits (2012).

6_{In particular, our interpretation is closer to the metaphysical reading. This does not}

mean that we fully agree with its ideas, since we take a completely different perspective, grounded on different assumptions. This will be clear in Section 3.2, and in Chapter 4.

7

The notions of form and structure are, in fact, part of the picture theory of meaning. We separated it into two Sections for the sake of the organization of the summary.

(7)

what as follows: What can be said at all can be said clearly; and whereof one cannot speak thereof one must be silent.

The book will, therefore, draw a limit to thinking, or rather – not to thinking, but to the expression of thoughts; for, in order to draw a limit to thinking we should have to be able to think both sides of this limit (we should therefore have to be able to think what cannot be thought).

The limit can, therefore, only be drawn in language and what lies on the other side of the limit will be simply nonsense.9

As we can see in this quote, Wittgenstein’s goal in TLP is to solve the problems of philosophy – not by analyzing them specifically, but rather by examining the causes from which they originate as a whole, and how this method relates to the logical structure of language. In order to do so, Wittgenstein aims to define the limit of meaningful language, so he can use it as a measure point to articulate what can be legitimately said. If the limit is properly established, then it can be stated that everything lying outside it is nonsense.10

Thus, by showing that the problems of philosophy are outside the limit, it can be concluded that they are not real problems – since they lack meaning. And it does not matter if they cannot be positively solved, because there is no need to solve them, after all. In other words, the solution offered by

9_{TLP Preface, p. 23. Note that there is some controversy regarding the ultimate goal of}

TLP. In this quote, it is made clear that one of the aims of TLP is to solve the problems of philosophy by setting the limit of meaningful language; however, most interpreters consider this a secondary goal. Wittgenstein was not clear about this issue. In his letter to Ficker in 1919 he claimed that “the point of the book is an ethical one.” This letter is cited in Stokhof (2002), pp. 5-6. It reveals that the goal of TLP is related to ethics. Furthermore, in a letter to Russell, Wittgenstein seems to imply that his main goal in TLP has to do with the relation between expression and inexpressibility, and that the limitation of language is just a means to this goal (see Section 2.1). We will not discuss this controversy, but we will provide our own interpretation regarding the goal of TLP in Chapter 4.

10

Wittgenstein makes a distinction in TLP between senseless (‘sinnlos’) and nonsensi-cal (‘unsinning’) propositions. Senseless propositions refer to those which are syntactinonsensi-cally correct but have no sense, because they do not represent a contingent situation – which, as we will see, is a necessary condition of meaningful language. Two representative cases of senseless propositions are tautologies and contradictions. Nonsensical propositions, on the other hand, have no sense because one or more of its constituents lack meaning – hence, they are not even syntactically correct, to start with. An example of nonsensical propo-sitions are the metaphysical propopropo-sitions, which concepts do not represent any object in reality. Both senseless and nonsensical propositions are similarly contrasted to proposi-tions with sense, insofar as both of them fail to represent, and hence, are illegitimate. In this paper, we will use the terms nonsense to refer to both nonsensical and senseless proposition, unless otherwise stated.

(8)

TLP to the problems of philosophy is a negative one. In this sense, it might be more accurate to say that the goal of Wittgenstein is not to solve the problems of philosophy, but rather to dissolve them. In any case, what is clear is that the notion of sense, and by opposition that of nonsense, are crucial in TLP. For this reason, the debate about the notion of nonsense in TLP cannot be regarded as accessory to the subject matter of the book; on the contrary, it can be argued that it deals with its central point.

1.1 The Picture Theory

The theoretical system that underlies the linguistic and semantic conceptions of TLP is known as the picture theory of meaning. As we will see in further Sections, the significance given to it varies greatly among different readings of TLP. Thus, while it is central for some interpretations, others regard it as a mere example of what cannot be done – set by Wittgenstein only to open the eyes of the reader and make her realize the futility of any theoretical enterprise. For the moment, we will not enter into the discussion about the centrality of the picture theory in TLP. In any case, what we can say is that it is what defines, at least from a theoretical angle, the limit of meaningfulness in the framework of TLP – and hence, it is particularly relevant for our summary and for the debate about nonsense, regardless of its significance to the philosophical insights of TLP.

The picture theory follows a representational conception of meaning. In other words, it defines meaning in relation to representation – i.e. the meaning of a sentence relates to that which it represents. Moreover, it is not simply a theory of language, but a broader theory of symbolic representation, that aims to explain how any symbolic system of representation works.11A picture – i.e. an element of a symbolic system – has sense only if it represents its object, and its meaning is that which it represents. On the contrary, a picture has no sense if it fails to represent its object – and hence, has no meaning attached to it.

Now, to clarify what is the conception of representation that Wittgen-stein has in mind – and, in particular, what is it that the picture must represent in order to have sense – we need to examine briefly the ontological side of the picture theory, since the explanation of the representational ob-ject of the picture depends on a particular ontological conception developed in TLP.12

11_{Note that the notion of ‘language’ in TLP is an abstract one – i.e., it refers to symbolic}

representation in general. Henceforth, we will use the term ‘language’ in this general sense.

(9)

onto-The first thing to note is that Wittgenstein follows in TLP an atomistic ontology, which is considered one of the main exponents of the philosophical doctrine known as logical atomism. It is an atomism because it identifies certain elements that are taken as single independent units and, from them, larger structures are formed. However, unlike other atomistic theories, the atomic elements in this ontology are not things but facts. To see it more clearly, let us have a look on the terminology used in TLP.

First, it is claimed that “the world is everything that is the case” (1), i.e. “the totality of facts” (1.1). A Fact – what is the case – is “the existence of atomic facts” (2). In other words, facts are formed by the combination of simpler facts. However, atomic facts are not considered the simplest entities in the framework, since they are built from the combination of even simpler entities: objects (2.01). Although they are the simplest entities, objects are not treated as atoms, because they cannot be the case on their own: they always depend on their connection to atomic facts (2.0122).

Importantly, facts can be actual or non-actual (1.21) and so can be atomic facts. In order to ease comprehension, we will use from now on the term ‘atomic fact’ to refer only to those that obtain, and the term state of affairs to refer to both actual and non-actual atomic facts; similarly, we will use the term ‘fact’ to refer only to those that obtain, and the term situation to refer to both actual and non-actual facts.13 Following this terminology, the atoms of this ontology are the states of affairs – which, combined, form situations .

Now, what defines the atoms in this system is not simplicity; it is log-ical independence (2.061), which is explained as follows:“[f]rom the exis-tence or non-exisexis-tence of an atomic fact we cannot infer the exisexis-tence or

logical system. Some interpreters prefer to separate the semantic aspects of picture theory from the ontology in which it is grounded – the Tractarian version of logical atomism. In our opinion, since the semantic and ontological aspects of the theory are directly con-nected in TLP, it is more useful to analyze them together, as two aspects of one and the same theory.

13

In the original version of TLP, Wittgenstein uses the terms ‘Sachlage’, ‘Sachverhalte’, and ‘Tatsache’ which, in the version we are using, are translated as state of affairs, atomic fact, and fact, respectively. The difference between ‘Sachlage’ and ‘Sachverhalte’ is related to actuality – i.e. an atomic fact is an actual state of affairs. As for the difference between ‘Sachverhalte’ and ‘Tatsache’, it has to do with complexity, as we mentioned. Wittgenstein uses the term ‘Sachlage’ to refer to both facts and atomic facts which are not necessarily actual. In this paper, we will follow the terminology used in Stokhof (2002), and we will further distinguish between states of affairs and situations – which are combinations of states of affairs. Hence, an atomic fact is an actual state of affairs, and a fact is an actual situation.

(10)

non-existence of another” (2.062).14 Note that logical independence makes states of affairs autonomous from one another, but it does not make them completely unconditioned, since they still depend on the particular way in which its objects are configured (2.0272). In logical atomism, objects – which “form the substance of the world” (2.021) – are fixed, i.e. they exist nec-essarily (2.027). Furthermore, they exist independently from what is the case (2.024).15 The difference between ‘independence from what is the case’ and ‘logical independence’ is crucial to understand logical atomism. Objects are neither actual nor non-actual by themselves, they merely exist as the substance of the world. They are the blocks that make up states of affairs. Actuality, on the other hand is tied to structure and configuration: what is actual or non-actual is the ways in which the objects are configured, i.e. states of affairs. Substance is independent from what is the case, but the defining aspect of the atom in this ontology is logical independence: the au-tonomy of entities in relation to their actuality, i.e. the impossibility to infer the actuality of an entity from another one. In other words, the atoms of the system must bear in themselves the possibility of being actual or non-actual. States of affairs bear this possibility but objects do not – and this is the reason why logical atomism is an ontology of facts and not of things.

Now, since logical independence is the defining aspect of atoms, this ontology is based on the notion of possibility, rather than existence or ac-tuality. Although the first propositions of TLP define ‘what is the case’, – and Wittgenstein builds the concepts of his ontology from there – what ultimately defines logical atomism is the contingent character of its atoms: states of affairs are the basic entities because they can be either actual or non-actual, which is a condition for their logical independence. Whether they obtain or not is not essential for their role as the atoms of the system. Thus, while the world is the totality of facts, as we mentioned, reality as a

14_{In other words, states of affairs are the atoms of the system because their actuality}

or non-actuality do not affect the actuality or non-actuality of other states of affairs in the configuration of a possible world. Situations, on the other hand, are not logically in-dependent, which means that a situation being actual or non-actual implies the actuality or non-actuality of other situations. For example, if the situation “It is raining in Amster-dam” is the case, then the situation “it is not raining in AmsterAmster-dam” cannot be the case. Hence, these two situations cannot be part of the same possible world.

15_{Note that there is a crucial difference between existence and actuality. The former,}

which refers to objects – and the substance of the world – is necessary, while the latter is contingent. As we will see, this is essential for the theses of the picture theory. In TLP 2.062, quoted above, it is said ‘existence’ and ‘non-existence’. In our opinion, this is confusing, as it actually refers, as can be deduced from the context, to ‘actuality’ and ‘non-actuality’.

(11)

whole gathers both actual and non-actual situations (2.06).

Reality is, then, the whole set of possible worlds – each one being a max-imally consistent set of situations – which also contains the actual world. Reality gathers all possible configurations of substance. These configurations are logical in nature, and for this reason, Wittgenstein also calls it logical space. Logical space determines compatibility or incompatibility of certain situations, the autonomy of states of affairs, the way some entities relate to simpler and more complex ones, and so on. This is the reason why this ontol-ogy is called logical atomism: because, ultimately, everything is determined by logical space. And this means that reality – and all entities contained in its different levels, namely states of affairs, situations, and possible worlds – are inherently logical. Or, more precisely, it is logic which organizes reality in terms of logical possibility.

Let us now go back to the notion of picture in order to answer the following question: What is it that a meaningful picture represents? TLP claims that what a picture must represent is a situation in logical space (2.11). Note that, since we are relying on an ontology of facts, we must understand a picture as a fact, not as a thing (2.141). In other words, a picture is a structured entity, and what it depicts must be a structured entity in turn. Hence, sense is described as a representational mapping between a complex symbolic production and an ontological structured possibility from the logical space. Whether the situation or state of affairs is actual or not only makes a difference with regards to the truth or falsity of the picture: it is true if it depicts a fact, and it is false if it depicts a situation that is not actual. But both true and false pictures are meaningful, as long as they manage to represent some possibility of the logical space. If a picture does not represent a situation or a state of affairs, it is neither true nor false: it has no sense.

The mapping between the picture and the situation is all there is to meaningful representation, which possibility is determined by the crucial isomorphic relation between language, thought and reality. First, thought it-self is interpreted as a pictorial system, since it is understood as “[t]he logical picture of the facts” (3), and “[t]he totality of true thoughts is a picture of the world” (3.01). Thought corresponds to reality, insofar as everything that is thinkable must be possible (3.02); in other words, thought is determined by the set of possibilities in the logical space. Furthermore, a thought is a proposition with sense (4), and a proposition is a thought “expressed per-ceptibly through the senses” (3.1). Propositions have a perceptible sign, the propositional sign, which expresses the though (3.12) and projects possible states of affairs (3.11). Hence, language – “[t]he totality of propositions”

(12)

(4.001) – is directly related to thought, and by extension, to reality.

To be more precise, it is meaningful language what is isomorphic to thought and reality. We mentioned that what determines the meaning of a picture is its correspondence with some situation in the logical space. Obviously, we can produce nonsensical sentences – if not, it would make no sense to search for the limits of sense. These nonsensical sentences, by definition, do not correspond to any possible world. But then, following what we said above, they cannot be thought. This means that only meaningful language counts as the expression of thought. Hence, the delimitation of meaning is crucially tied to the isomorphic relation between properly used language, thought, and the set of possible worlds: for a sentence to have meaning it has to express a thought and project the possible situation that this thought relates to – being it actual or not. Through this expression and this projection, the sentence acquires its meaning in the representation of the situation.

1.2 Form and Structure

Now, we come back to logic, which, of course, plays a central role in the Trac-tarian definition of meaning. Indeed, it is logic what underlies the isomorphic relation between language thought and reality: it is what these three aspects have necessarily in common. This is crucial, because what makes picturing possible, in TLP, is the connection between the picture and the depicted – a connection that is essentially logical. Wittgenstein postulates that there is only one system of logic, which pervades and connects reality, thought and language.16, For this reason, to say that some situation is impossible, that some thought is inconceivable, or that some sentence is nonsensical, is equivalent to saying that this situation, thought or sentence are illogical.

In relation to ontology, we mentioned that what defines an entity in TLP is not existence but possibility. Also, we mentioned that ontological possibility in TLP is logical in nature, since it is determined by logical space. Logical space is not only the collection of all possibilities: it is also what organizes them, i.e. what gives a structure to the totality of possibilities. This structure has a particular form – called logical form – which is ultimately what connects different possibilities and allow representation. The analysis of the notions of form and structure bring us the final clue to understand the functioning of the picture theory.

16_{The absolutism of logic is not explicitly stated in TLP, but rather, it can be deduced}

(13)

Form, in TLP, is what characterize a certain entity in relation to its par-ticular set of possibilities. The form of some entity is the set of possibilities that determines its ability to build larger blocks. For example, the form of an objects is the set of possible ways in which it can combine with other objects to build states of affairs (2.0141). On the other hand, the form of an entity also defines which elements can combine to build it. Hence, for instance, the form of states of affairs is not only their possibility to build situations, but also their possibility to be built from objects. The form of an entity, then, connects it – in terms of its possibilities – to the form of the elements that can build it, and to the form of the larger entities that can be built from the entity in question.

This characterization of the notion of form applies to all the ontological distinctions mentioned above. It also applies to pictures – after all, a picture is just a situation which is used to represent another situation. The form of the picture is the way in which the elements that can possibly build it may be combined – in relation to their own combinatorial possibilities – and the way it can combine with other pictures to build larger, more complex ones. For example, the form of a sentence determines, on the one hand, the kinds of words that can be combined to build it, and on the other hand, the ways in which the sentence in question can combine with others to generate more complex sentences. Now, inasmuch as it is a situation depicting other situation, the form of the picture is also a pictorial form. This notion defines the possibility that the elements of the picture are combined in the same way as the elements of the situation it depicts (2.151) – a possibility which requires that the picture and the situation it depicts share, at some level, the same form. And this is ultimately achieved thanks to both of them sharing logical form.

Logical form is the form of reality – i.e., the form of the logical space (2.18).17 It is, then, the possibility of all different configurations of possibil-ity itself, including the actualpossibil-ity or non-actualpossibil-ity of situations and states of affairs. In other words, it is the most general form: it connects all entities and defines their formal particularities and possible combinations. Hence, logi-cal form is what ultimately makes possible representation – it is the most fundamental connection between pictures and situations, and it is what de-termines the formal particularities that make a specific picture suitable to

17_{“What every picture, of whatever form, must have in common with reality in order}

to be able to represent it at all – rightly or falsely – is the logical form, that is, the form of reality” (2.18). Reality, as we mentioned, is the totality of all possible situations. Since all situations can be deconstructed in all possible states of affairs and objects – i.e. all possibilities – reality and logical space are equivalent.

(14)

depict a specific situation. For this reason, “[e]very picture is also a logi-cal picture” (2.182). And this means that representation itself is logilogi-cal in nature.

So far we have talked about form. However, we must take into account that this notion only characterizes an entity in terms of its possibilities; it does not define it in its singularity – since entities of the same kind share the same form. In this sense, form is necessary, but not sufficient for representa-tion. Hence, Wittgenstein introduces the notion of structure. The structure of an entity refers to both its form and the specific set of elements of which it is composed. It is, then, what ultimately defines a particular entity. More-over, abstracting from the particular elements that build an entity, its form is revealed – and replacing its elements with others with the same form, we get a different entity that shares the form with it. Through the notion of structure we get to the general definition of form: “form is the possibility of structure” (2.033).

Meaning – in the picture theory – requires that the picture and the situation it depicts share the same form (2.17). But not only that: it also requires a pictorial relationship between the picture and the situation, i.e. it also requires that the elements of the picture do represent the elements of the situation depicted (2.1514).18Both requirements are ultimately satisfied, as we mentioned, thanks to all situations having in common logical form, which makes possible pictorial representation, and thanks to all possibilities being necessarily part of the logical space, which connects all ontological levels under a general logical structure, allowing pictorial relationship.

In conclusion, the picture theory provides a characterization of meaning that defines it in terms of representation, along with an ontological system – grounded on possibility and on the notions of form and structure – that describes how representation work. With these resources, we can now tackle the problem of the limits of meaningful language. Language is the totality of propositions. A proposition is a picture; it is meaningful if it depicts a situation. Situations can be either actual or non-actual. In fact, it is this contingency what characterizes them. On the one hand, situations cannot be impossible, insofar as they pertain to the logical space, which is the general structure of possibility. A situation is, by definition, a possible case; an impossible case is inconceivable and illogical, so it does not count as

18_{“The representing relation consists of the co-ordinations of the elements of the picture}

and the things” (2.1514). See also TLP 4.014: “The gramophone record, the musical thought, the score, the waves of sound, all stand to one another in that pictorial internal relation, which holds between language and the world. To all of them the logical structure is common.”

(15)

a situation in Tractarian terms. On the other hand, a situation cannot be necessary. A fact is just a situation that obtains, but there is nothing intrinsic to the fact that makes it necessarily actual. The only thing intrinsic to any situation, actual or not, is its structure – and by abstraction its form – but nothing in the structure or the form of an entity determines necessarily its actuality: they only define the elements that build it, the way in which these elements are combined, and the possibilities that this particular combination opens up.

Importantly, in TLP, logical necessity is the only kind of necessity (6.37). Structure and form are the logical determination of a situation: they are what locate it in the logical space; hence, they are necessary in turn. How-ever, the situation itself cannot be reduced to the necessity of logic, since it can be actual or non actual, and nothing in its logical form determines its actuality. In conclusion, every meaningful picture – and, by extension, ev-ery meaningful proposition – must represent a situation, which is essentially contingent. The limit of language, then, is contingency. In other words: we cannot say anything meaningful that is impossible, nor can we say anything meaningful that is necessary – including anything logical.

1.3 The Distinction Between Saying and Showing

The conception of meaning that follows from the picture theory is absolute. There are no exceptions: meaning refers exclusively to the representation of a contingent situation by a picture – which is another contingent situation that shares its form and is connected to it by a pictorial relationship. This conception defines clearly the limit of representation. The use of language for anything different than the representation of a contingent situation results in a picture with no sense. In particular, it can be a nonsensical picture – if one or more of its constituents lack meaning – or a senseless proposition – if all its constituents are meaningful but it does not represent a contingent situation. Senseless propositions are logical propositions, namely, tautologies (6.1).19 Tautologies are “part of the symbolism” (4.4611) – i.e., they are syntactically correct. However, they do not represent anything, and hence, they are not pictures (4.462). But they show something about the logical form that underlies representation and reality (6.12).20

19

Cf. note 10. Note that contradictions are also logical propositions (6.1202).

20_{For example, the logical proposition ¬(p ∧ ¬p), being a tautology, shows that the}

propositions p and ¬p are contradictory, i.e., it shows that these propositions cannot obtain at the same time. Or, in other words, that the situations they depict cannot be together in the same world – which is, as we mentioned, a maximally consistent set of

(16)

The notion of showing – and, by extension, the distinction between ‘what can be said’ and ‘what shows itself’ – is one of the most extensively treated topics in TLP. It is also essential to understand the debate on nonsense, since it can be interpreted as an argument for a non-linguistic way of com-prehension beyond the notions of meaning and truth. On the other hand, some readings seem to disregard the idea altogether. Again, we will leave the discussion about its relevance for further Sections, and we will focus on what Wittgenstein actually says about it in TLP.21

First, let us see what the distinction is meant to describe. ‘What can be said’ refers to everything that can be pictured. In other words, it refers to everything that lies within the limits of sense, as defined by the picture the-ory – i.e., the representation of contingent situations. What shows itself, on the other hand, refers to everything that cannot be pictured but is neverthe-less revealed, in some way. It is not clear, in TLP, if this revelation implies some special kind of comprehension, or if it is just something implicit but impossible to comprehend in any way – hence the controversy, or at least one side of it. What is clear is that it refers to something that cannot be represented, and thus, something that is not governed by the principles of meaning developed in TLP.

Moreover, the distinction is meant to describe a dichotomy, following the claim that “what can be shown cannot be said” (4.1212) – and the idea, not directly expressed but implicit through TLP, that what is said differs from what shows itself.22 However, this dichotomy must not be understood as a separation between meaningful and meaningless language. Rather, it refers to two opposed sides of every picture. In other words, every picture says – represents – something and, at the same time, it shows something else. In particular, a picture “shows its sense” – i.e “shows how things stand, if it is true” – and “it says, that they do so stand” (4.022). Now, to claim that the picture shows ‘how things stand’ means that ‘what shows itself’ in the picture is, basically, the logical space – in which the situation depicted by the picture is structured.23_{To see it more clearly, we can quote TLP 4.121:}

possible situations.

21

We must mention, however, that this particular topic is specially dependent on inter-pretation, i.e., is difficult to read at face value. We will provide a reading as impartial as possible, in order to avoid taking sides in the debate on nonsense.

22

Cf. TLP 3.262 and TLP 4.121.

23_{The idea is that the picture shows the logical space, insofar as the latter defines a}

particular arrangement of situations in a possible world. It says that this situation is actual, which can be true or false depending on whether the situation actually obtains or does not obtain, respectively.

(17)

4.121 Propositions cannot represent the logical form: this mirrors itself in the propo-sitions.

That which mirrors itself in language, language cannot represent. That which expresses itself in language, we cannot express by language. The propositions show the logical form of reality.

They exhibit it.

Logic – despite being necessary, and thus meaningless – shows itself in language. As we mentioned, logic, in TLP, is the ground that connects a picture with its depicted situation. It is only because the picture and the situation share the same logical form that the former can represent the latter. Even if it is something that cannot be said – since it is not a situation itself – this logical connection between the picture and the situation is mirrored in language. ‘What shows itself’, then, is the one and only logic that underlies reality and the possibility of its representation.24

Now, recall that logic is the only necessity there is (6.37) In other words, everything that is necessary is logical. However, we must not understand this as a limitation of necessity, but rather as an extension of the notion of logic, to cover everything that is necessary in reality – and in the representation of reality. It is clear in the case of mathematics (6.2): in TLP, a mathematical proposition, like a logical proposition, is senseless. And just like tautologies, mathematical equations show the logic of the world (6.22).

Although it is not so clear, it can also be applied to other frameworks that display some kind of necessity about reality – or about the representation of reality. For example, Wittgenstein claims that “what solipsism means, is quite correct, only it cannot be said, but it shows itself” (5.62). The point is that “[t]he limits of my language mean the limits of my world” (5.6), and these limits are, basically, the limits of logic (5.61). In other words, what solipsism means is something that is necessary for the act of representation – it is logical – and hence, it is mirrored in language. It cannot be said, but it shows itself. Another example is related to the principles of science. Again, these principles are necessary, i.e. they are logical; hence, they cannot be said, but show themselves.25

Finally, we get to the notion of the mystical.26Let us see this one in more

24_{Another example is found in TLP 4.122, where Wittgenstein claims that internal}

properties and relations – i.e., the logical form – of situations cannot be said, but are shown in propositions.

25

In particular, Wittgenstein mentions the law of causality (6.36)

26

We will not address this notion in detail. For a detailed analysis, cf. Arnswald (2009) Atkinson (2009), and Sontag (1995).

(18)

detail. We will quote the four statements in which Wittgenstein explicitly addresses this notion:

6.432 How the world is, is completely indifferent for what is higher. God does not reveal himself in the world.

6.44 Not how the world is, is the mystical, but that it is.

6.45 The contemplation of the world sub specie aeterni is its contemplation as a limited whole.

The feeling of the world as a limited whole is the mystical feeling. 6.522 There is indeed the inexpressible. This shows itself; it is the mystical.

The idea seems to be the following: the mystical is the feeling of the world as a limited whole, i.e. of reality along its own limits. These feeling does not refer to the world as a particular arrangement of situations, but rather to the feeling that reality holds – “not how the world is but that it is”. ‘How the world is’ is something contingent, but ‘that it is’ is necessary. In other words, the mystical feeling is also inherently logical – and this is clear from the fact that the mystical is the contemplation of the world as a limited whole, and the limit of reality is the same as the limit of its representation: logic. The mystical is the inexpressible, i.e., it cannot be said; but it shows itself as the limit of the world – and the limit of the world is logic.

We will come back to this topic in Section 3.1, where we will present our own interpretation of TLP. For the moment, let us finish the summary with an outline about the role of philosophy in TLP.

1.4 The Role of Philosophy

Our task in this final part of our summary will be to build a notion of philosophy from what is actually written in TLP about it. One of the main topics of the debate has to do with the philosophical status of TLP - i.e., whether TLP can be considered a philosophical work or, on the contrary, a manifest against philosophy. We will analyze these two perspectives in Chapters 2 and 3 and we will give our own answer in Chapter 4. For the moment, the idea is simply to present what TLP says about philosophy, and how it connects with the picture theory of meaning.

Recall that, as stated in the Preface, one of the aims of TLP is to show that the traditional problems of philosophy can be solved by reducing them to a more general metalinguistic problem: that of the limits of language. We find another claim in the same line of argumentation in TLP 4.003:

(19)

4.003 Most propositions and questions, that have been written about philosoph-ical matters, are not false, but [nonsensphilosoph-ical]. We cannot, therefore, answer questions of this kind at all, but only state their senselessness. Most ques-tions and proposiques-tions of the philosophers result from the fact that we do not understand the logic of our language.

(They are of the same kind as the question whether the Good is more or less identical than the Beautiful.)

And so it is not to be wondered at that the deepest problems are really no problems.27

The idea is the following: the problem of philosophical propositions lies in their lack of sense. This lack of sense is derived from a misunderstanding of the logical functioning of language.28In connection with the picture theory of meaning, we can say that philosophical propositions are nonsensical because they fail to represent a situation. They use concepts that do not correspond to anything in reality, and thus, they lack meaning. In other words, they are not true or false, because they lack meaningful content, to start with. Hence, the problems that these propositions exhibit cannot be solved because they are not problems in the first place – they can only be proven nonsensical, and thus be dissolved. Undoubtedly, this statement – in addition to the claims in the Preface – present a negative view on philosophy, and can serve as an argument for the idea that TLP proposes a rejection of the philosophical framework, insofar as it has no sense and it is simply the product of a confusion with regards to the functioning of representation.

However, it must be noted that there are other statements in TLP which talk about philosophy in a more positive way. First, in relation to the object of philosophy Wittgenstein writes the following:

4.112 The object of philosophy is the logical clarification of thoughts. Philosophy is not a theory but an activity.

A philosophical work consists essentially of elucidations.

The result of philosophy is not a number of “philosophical propositions”, but to make propositions clear.

Philosophy should make clear and delimit sharply the thoughts which oth-erwise are, as it were, opaque and blurred.

27

We changed the term ‘senseless’ used in Ogden’s translation, for the more correct ‘nonsensical’. Regarding this issue, cf. note 33.

28

Example in 4.128; also, example of common confusions, which are not limited to philosophy, but on the contrary have their origin in ordinary language (3.323 - 3.324, 4.122).

(20)

The function of philosophy is not to examine some abstract or tran-scendent problems, but rather to make clear which thoughts are logically acceptable and which ones are not. It is implied that, for some particular thoughts, it is not apparent at first whether they are logically licit or not; they are ‘opaque and blurred”, and some analysis is necessary to determine whether they comply with the rules of logic. The task of philosophy is, then, to elucidate the legitimacy of these thoughts.29 But, in order to do this, it is necessary to go from the thoughts to the propositions that express them, since it is only possible to establish the limit between meaningful and mean-ingless propositions.30 Hence, the problem that philosophy aims to solve is the metalinguistic problem of the legitimacy of propositions, i.e., the funda-mental problem for which the apparatus of the picture theory is built.

However, Wittgenstein makes clear that philosophy is not a theory which results in a set of propositions, but rather it refers to the bare activity of clarification of propositions. What Wittgenstein proposes is a new way to understand philosophy: since it cannot be regarded as a particular set of propositions, a philosophical work must be understood as a production in which the activity of clarification of thoughts is performed. Hence, if TLP can be considered a philosophical work, it is because there is a process of logical clarification in it. But, more importantly, TLP is mostly a meta-philosophical work, since it is devoted to present the theoretical ground from which the philosophical practice of elucidation can be executed. The picture theory, then, is not philosophy – since the latter is not a theory – but rather it is meta-philosophy, insofar as it provides the theoretical structure that makes philosophical practice possible.

We have, then, a positive notion and a negative notion of philosophy. In order to avoid confusions, let us use the term elucidation to refer to the positive notion of philosophy in TLP, and metaphysics to refer to the neg-ative one. ‘Metaphysics’, then, is the totality of philosophical propositions, which are found to be nonsensical by ‘elucidation’. The claim is that there are no proper philosophical propositions whatsoever. ‘Elucidation’, on the other hand, is a practice, not a linguistic framework: it is not a set of propo-sitions but the activity of logical clarification of propopropo-sitions. In fact, one of its crucial roles is to unveil the nonsensical character of ‘metaphysics’.31

29

We will analyze in more detail the notion of elucidation in Section 4.2. For the moment, we can just take it in its more common meaning, as a synonym of ‘clarification’.

30

Cf. TLP Preface, p. 23.

31_{Recall that one of the goals of TLP is to solve the problems of philosophy, as stated in}

the TLP Preface, p. 23.). The problems of philosophy are reduced to a problem of language. Elucidation is, then, what allows us to solve the problems of philosophy, because it clarifies

(21)

Philosophy, clearly, “is not one of the natural sciences” (4.111),32and, in fact, it is neither true nor false. It is nonsensical, if we refer to metaphysics, and it is not even propositional, if we talk about elucidation. Nevertheless, philosophy, understood as elucidation, “limits the disputable sphere of nat-ural science” (4.113), since it is the activity that carries out the delimitation of sense established by the picture theory.

Following this ideas, now that we are clear about the positive and neg-ative interpretations of philosophy, we must quote an important statement at the end of TLP in which the method of philosophy is stated:

6.53 The right method of philosophy would be this. To say nothing except what can be said, i.e. the propositions of natural science, i.e. something that has nothing to do with philosophy: and then always, when someone else wished to say something metaphysical, to demonstrate to him that he had given no meaning to certain signs in his propositions. This method would be unsat-isfying to the other – he would not have the feeling that we were teaching him philosophy – but it would be the only strictly correct method.

In this statement, Wittgenstein makes clear that the proper way to pre-form philosophy is elucidation – i.e., clarification of propositions. He also states that the problem of metaphysics – what makes its propositions non-sensical – is that certain signs lack meaning. Furthermore, it is stated that we must remain within the limits of what can be said, the limits of sense (7). Note that, since elucidation is not propositional, this cannot be applied to philosophy itself – “what can be said [...] has nothing to do with philosophy” – but rather it must be seen as a constitutive tenet of TLP as a whole. In any case, this tenet shows why the picture theory and philosophical practice are necessary: we must be able to elucidate the legitimacy of propositions in order to avoid saying something that cannot be said.

language and reveals the lack of sense of philosophical propositions.

32

Note that natural science is used in a more general way in TLP than the ordinary notion: it is not simply a particular set of propositions with a common content, but rather “the totality of true propositions” (4.11). In other words, every proper picture of a fact is considered natural science. In other words, science is in TLP the language of contingency – i.e., the only proper meaningful language.

(22)

2 The Debate on Nonsense. Two Attempts to Solve

the Paradox of TLP

If we have a look on the summary just presented, we can see that there is an important issue affecting the consistency of the picture theory of meaning. The results of the theory make it clear that sense is tied to contingency. However, the theory itself is not contingent – or, more precisely, it does not represent contingent situations. What it aims to represent is the ‘logic of the world’– both the ground of reality and the ground of representation. But the ‘logic of the world’ is not a situation, and it cannot be represented. In other words, the picture theory, as a semantic and ontological framework, has no sense. This is a crucial issue, because it implies that the theory undermines itself : it is nonsensical only if we accept its definition of sense, in the first place; but if it is indeed nonsensical, then there is no reason to accept this definition, since it lacks sense. We will refer to this essential feature of the picture theory as the paradox of TLP.

Our claim is that the paradox is not something accidental to TLP. Wittgenstein was well aware of the self-refuting character of the picture theory, as proven by TLP 6.54:

6.54 My propositions are elucidatory in this way: he who understands me finally recognizes them as [nonsensical], when he has climbed out through them, on them, over them. (He must so to speak throw away the ladder, after he has climbed up on it.)

He must surmount these propositions; then he sees the world rightly.33

This statement makes clear that the proper understanding of TLP im-plies the recognition of the nonsensical character of its propositions – i.e., the

33_{In the version of TLP that we are using, the term used is ‘senseless’ instead of}

‘non-sensical’. Note that in the original version, the term used is ‘unssining’, which we have translated as nonsensical – in contrast to ‘sinnlos’, which we translate as senseless (cf. note 10). This confusion is related to the fact that Ogden does not differentiate between the two notions, which is clearly an important shortcoming of his translation. This is relevant because the consideration of TLP as a nonsensical work – instead of a senseless one – has crucial consequences for the interpretation of the book, as we will see further on. Note that it makes more sense to interpret the propositions of TLP as nonsensical; as Roy Brand points out in the following passage: “[the propositions of TLP] are not merely logical propositions (i.e. tautologies or contradictions) but rather they attempt to speak about those propositions. The liar’s paradox, insofar as it is a mere contradiction, is not nonsense but senseless, but insofar as it is used to make a point about logical contradictions it is nonsense.” Cf. Brand (2004), note 35.

(23)

recognition of the paradoxical character of the picture theory.34Importantly, he also claims that, despite their nonsensical character, the propositions of TLP allow the reader to ‘see the world rightly’, insofar as they are ‘elucida-tory’. We will come back to these ideas in Chapter 4, where we will present our own interpretation. For the moment, we want to note that the paradox of TLP is not simply a problematic result of the picture theory, but rather, it is something essential to understand what Wittgenstein wanted to show in TLP.

In this Chapter we will present our view on the debate on nonsense, and the two opposed readings that characterize it – which we will call the therapeutic reading and the metaphysical reading.35We consider this debate crucial for our interpretation of the book, particularly in relation to its paradoxical character. The two readings in the debate have one thing in common that makes them different from the previous interpretations of TLP: both of them take the paradox of TLP as a central aspect of their – otherwise confronted – interpretations. We call it the debate on nonsense because the two readings organize their interpretations around two opposed analyses of the notion of nonsense.36 These analyses suggest two different ways to solve the paradox. In other words, through the notion of nonsense, each reading aims to avoid the inconsistencies produced by the paradox, in order to provide an interpretation in which TLP has sense.

Why is the notion of nonsense relevant for the paradox of TLP? First, it must be clear that, at least within the framework of TLP, there is no

34

We are assuming here that the recognition of the nonsensical character of the propo-sitions of TLP amounts to the recognition of the paradox. This is a plausible assumption, insofar as the nonsensical character of these propositions follows from the implementation of the theses of the picture theory to itself. Hence, to acknowledge the nonsensical char-acter of the propositions of TLP we must first accept the theory, but to accept the theory implies its rejection as a nonsensical construction at the same time. Hence the paradox.

35_{The two positions of the debate have been named in many different ways. We chose}

these terms because they reflect clearly, in our opinion, the relation between the debate and the paradox. As we will see further on, the metaphysical reading is based on the assumption that TLP is, mainly, a philosophical work, while the therapeutic reading interprets TLP as a therapy against metaphysics.

36_{We must note that, although the positions are clear, the debate itself is open to}

dif-ferent interpretations. In fact, to focus on the notion of nonsense is already a particular way of presenting the debate. Other commentators have analyzed it from different start-ing points, such as the distinction between ‘what can be said’ and ‘what shows itself’, the validity of philosophical theories in general, or the role of TLP and its position in Wittgenstein’s philosophy as a whole. In any case, we can see from the structure of TLP that these topics are closely related to one another. For this reason, focusing on a partic-ular topic only affects the emphasis given to it in the contrast between the two readings of the debate.

(24)

such thing as an abstract notion of ‘nonsense’, opposed to an abstract no-tion of ‘sense’ – like two realms divided by the limits of meaningfulness. On the contrary, these notions only function in the context of the picture the-ory, only when they are applied either to failed attempts of representation or to successful ones, respectively. Hence, the picture theory is nonsensi-cal only insofar as its propositions are nonsensinonsensi-cal. Now, as we mentioned, the paradox is a consequence of the self-falsifiability of the picture theory of meaning: if we accept the theory, we must conclude, by its own results, that its propositions are nonsensical. This means that the characterization of nonsense in TLP is nonsensical itself, since it can only be defined through the propositions of the picture theory of meaning. In other words, the no-tion of nonsense can only be demarcated by nonsensical proposino-tions. But in order to define what it is a nonsensical proposition we need a theory of meaning that tells us what it is to have sense. Hence, the notion of nonsense is specially relevant because it reflects, in a particular and definite way, the paradoxical character of the picture theory: a nonsensical theory of sense.

The point of disagreement between the two readings of the debate is related to the significance of nonsense – both its relevance for the under-standing of TLP, and the way it signifies in terms of its informative content. A significant interpretation of nonsense – one in which it is informative – allows us to preserve the theory from total self-dissolution, because even if we accept that the theory is nonsensical by its own standards, it still shows something which is in fact more fundamental than what can be meaningfully said. Hence, there is no paradox after all, just a back a forth game between sense and nonsense. On the other hand, if we want to preserve a strict notion of nonsense, we can only do it by jettisoning the theory. This second option requires a complete reinterpretation of TLP, in which the goal defined in the Preface is not actually the main objective of Wittgenstein – or at least it is not accomplished by means of the picture theory. The advantage of this interpretation is that it allows us to comply with the fundamental tenet of TLP: “whereof one cannot speak, thereof one must be silent” (7). These two positions define the metaphysical reading and the therapeutic reading, respectively.

Before we proceed to the presentation of the two readings, we must say a word about the interpretative ground of the debate on nonsense. The debate emerges as a response to the traditional interpretation of TLP – also called the orthodox interpretation.37 _{The traditional interpretation focuses}

37

The traditional interpretation is a general reading of TLP based on Russell’s introduc-tion to the first translaintroduc-tion of TLP, and on the analyses by authors such as Frank Plumton

(25)

on the formulation and the results of the picture theory. It holds that the main objective of TLP is to offer a semantic theory of representation, with logic as the ground of both language and reality. Other themes of the book, such as the reformulation of the method of philosophy, or the rejection of metaphysics, are just considered particular outcomes of the picture theory.

The traditional interpretation is not specially interested in the paradox of TLP. However, it stills views the distinction between ‘what can be said’ and ‘what shows itself” as a crucial idea of TLP. As Russell writes in the introduction to TLP:38

In order that a certain sentence should assert a certain fact there must, how-ever the language may be constructed, be something in common between the structure of the sentence and the structure of the fact. This is perhaps the most fundamental thesis of Mr Wittgenstein’s theory. That which has to be in common between the sentence and the fact cannot, so he contends, be itself in turn said in language. It can, in his phraseology, only be shown, not said, for whatever we may say will still need to have the same structure.

For traditional interpreters, this distinction is just another consequence of the picture theory. However, it is crucial to explain the problematic conclu-sion expressed in 6.54. The propositions of the picture theory are nonsensical because they describe the ‘logic of the world’ – and they conclude that noth-ing can be said about it. But the logic of the world is still shown – not only through propositions with sense, but also through nonsensical ones, like the ones of the picture theory. Thus, by making the distinction a fundamental thesis of TLP, traditional interpreters are able to carry on with the analysis of the picture theory, even if it is made of nonsensical propositions. In any case, this solution does not tackle the fundamental problem of the paradox of TLP; it just sets it aside.

The two readings of the debate are formed by authors that either object or defend this traditional view. The insufficient treatment of the paradox of TLP by traditional interpreters is the seed of the debate. The therapeutic reading attacks the notion of nonsense presented by the traditional interpre-tation; they argue that the paradox of TLP makes this notion of nonsense inconsistent in relation to the results of the picture theory. The metaphysical reading, on the other hand, takes into account the criticism of therapeutic readers, and aims to adapt the traditional interpretation of the notion of

Ramsey, Norman Malcolm and Elisabeth Anscombe, among others. This interpretation is one of the most influential and one of the most extended in time. Its advantage is that its contributors were all personally – if not philosophically – close to Wittgenstein.

38

(26)

nonsense to the paradox, in order to defend it from the objections of the therapeutic reading.

We will now examine these two readings in more detail in the following Sections. We will start by the metaphysical reading because, despite the fact that it is a reaction against the criticism of the therapeutic reading, it connects directly with the traditional interpretation – and hence, it will help us understand what is the target of the criticism of therapeutic reading. In Chapter 2 we will introduce the two readings, but we will leave our evaluation of these readings for Chapter 3.

2.1 Metaphysical Reading

To introduce the metaphysical reading, we will start by presenting its most representative analysis of the notion of nonsense, which is found in Anscombe (1959).

This book is a detailed analysis of TLP. Published in 1959, almost 40 years after TLP came to light, it can be considered one of the latest exam-ples of the traditional interpretation. The book consists of 13 Chapters; the notion of nonsense is analyzed in the last one. The rest of the book is devoted mostly to the logical and semantic insights of TLP. As we mentioned, this is one of the main features of the traditional interpretation: the emphasis is set on the picture theory of meaning and its views on logic. However, in the beginning of Chapter 13, Anscombe quotes a letter from Wittgenstein to Russell which warns the latter about the mistaken focus of the traditional interpretation. We will reproduce it here:

Now I’m afraid you haven’t really got hold of my main contention, to which the whole business of logical propositions is only corollary. The main point is the theory of what can be expressed by propositions i.e. by language (and, what comes to the same, what can be thought) and what cannot be expressed by propositions, but only shown; which, I believe, is the cardinal problem of philosophy.39

This quote expresses a point that we mentioned above: the centrality in TLP of the distinction between ‘what can be said’ and ‘what shows itself’. Wittgenstein’s comment in the letter relates to the understanding of TLP as a whole, and pinpoints a problem that is characteristic of the traditional interpretation. The distinction is not, as traditional interpreters have it, a mere result of the picture theory. Rather, the theory is a corollary of the distinction. In other words, the ultimate goal of TLP is not to define

(27)

the functioning of representation in order to set the limit of sense. The distinction between ‘what can be said’ and ‘what shows itself’– which is a philosophical distinction – is the core of TLP, and the limit of sense is just a necessary tool to develop it. This implies that TLP is not primarily a logical-semantic work: it is a philosophical work which contains – and is built over – a logical-semantic theory.

The quote of Wittgenstein’s letter to Russell reveals an idea that later became the main feature of the metaphysical reading: the characterization of TLP as a philosophical work, illustrated by the distinction between ‘what can be said’ and ‘what shows itself’. Hence, it can be said that the meta-physical reading develops the traditional line of interpretation by solving this essential problem that has to do with the characterization of TLP – which involves the description of the interaction of its different themes, and the definition of its ultimate goal. This is also the reason why it is called ‘metaphysical reading’ of TLP.

The relevance of Chapter 13 of Anscombe’s book does not end here. It also presents the analysis of nonsense that is at the center of the debate,as we mentioned. Let us start with a crucial claim of Anscombe just a few paragraphs below the quote of Wittgenstein’s letter. Anscombe writes the following:

[...] An important part is played in the Tractatus by the things which, though they cannot be ‘said’, are yet ‘shewn’ or ‘displayed’. That is to say: it would be right to call them ‘true’ if, per impossibile they could be said; in fact they cannot be called true, since they cannot be said, but ‘can be shewn’, or ‘are exhibited’, in the propositions saying the various things that can be said.40

It is important to note that the notion of truth is strictly defined by the picture theory. Truth and falsity are confined in TLP to the sphere of what can be said. As we know, only representation conveys truth – when the situation represented is actual – and falsity – when the situation depicted is non-actual. In other words, there is no space for truth in the sphere of ‘what shows itself’. A nonsensical proposition is, strictly speaking, a proposition that is neither true nor false. Anscombe is well aware of this point, central to the picture theory. To understand what Anscombe is trying to express in this quote, we must look, again, at TLP 5.62. Let us recall what is claimed in this statement: “[...] what solipsism means, is quite correct, only it cannot be said, but it shows itself”. Hence, what ‘would be true’ if it ‘could be said’ in Anscombe’s quote is equivalent to what is quite correct even if it cannot

40

Ibid., p. 162. Note that Anscombe uses the term ‘shewn’ instead of the more commonly used shown.

(28)

be said, in 5.62. The idea is the same: the information we receive from ‘what shows itself’ is something that cannot be said – but it cannot be negated neither. It is right in a way that goes beyond the limit of propositional truth. Anscombe’s idea of ‘what would be true if it could be said’ has been later defined through the notion of ineffable truth.41 It is this notion that defines most clearly the line of interpretation of the metaphysical reading. As we mentioned, the metaphysical reading is characterized by taking into account the philosophical point of TLP as a central aspect of the book – and even portraying TLP primarily as a philosophical work. This philosophical point with which Wittgenstein is concerned in TLP is the tension between ex-pression and the inexpressible, represented through the distinction between ‘what can be said’ and ‘what shows itself’.

The centrality of the distinction allows metaphysical readers to claim that the paradox of TLP is only apparent. The idea that ‘what cannot be said’ is shown, implies that it is possible to grasp the inexpressible in some way. This is, for metaphysical readers, the significance of TLP 6.54: the reader is able to ‘see the world rightly’ after ‘climbing up the ladder’, once he has ‘thrown away the ladder’ – i.e. once he has realized that the propositions that have brought her to this point are in fact nonsensical. The notion of ineffable truth aims to this idea: the view that emerges from the picture theory is correct despite being inexpressible. It is ‘like a truth’, but it cannot be directly expressed through language. It lacks meaning, but it is correct nonetheless.

Now, since the propositions of TLP are supposed to show some ineffable truth, and since they are nonsensical propositions, it follows that nonsensical propositions do, in fact, show something. It is important to mention this, because it is not something obvious from the reading of TLP – i.e., it is a distinctive mark of the metaphysical reading. In any case, this interpretation is essential for the solution of the paradox proposed by metaphysical readers. However, the claim that nonsensical propositions show some ineffable truth is problematic. A nonsensical proposition can be any kind of illegitimate proposition – even a random string of words – so it is difficult to hold that all nonsensical propositions show something. The strategy of metaphysical readers is simple: they distinguish between mere gibberish and illuminating nonsense. The former applies to constructions that do no ‘attempt’ to say anything, and cannot even be considered propositions. At most, they do

re-41

Peter Hacker, the most representative figure of the metaphysical reading along with Anscombe, is the interpreter who first used this notion, which became a distinctive notion of the metaphysical reading. Cf. Hacker (2000).

The Paradoxical Reading of The Tractatus