What Wittgenstein did and what He Said: Commentary to David Hommen’s ‘Wittgenstein, Ordinary Language, and Poeticity’

1 Introduction

In his article ‘Wittgenstein, Ordinary Language and Poeticity’, David Hommen addresses the apparent inconsistency (315^[1]) or contravening tendencies (313) in the role that ordinary language or ‘language of everyday life’ (313) plays in the later Wittgenstein’s works. This is a central topic in the interpretation of Wittgenstein’s works as it relates to both the origin of the problems that Wittgenstein concerns himself with and the method by which elucidations or solutions to these should be sought.

Hommen’s article is a recent contribution to this topic which started in the 1950s. It was initially a question of the later Wittgenstein’s role in a dominant, post-war philosophical orientation originating in Britain, the so-called ‘ordinary language philosophy’ or ‘Oxford philosophy’.^[2] During the 1970s and 1980s, two directions of interpretation of Wittgenstein’s later philosophy gradually evolved: The grammatical approach with G.P Baker and P.M.S Hacker as prominent proponents^[3] and the therapeutic approach with Stanley Cavell as an early contributor.^[4] There are multiple variations of these ‘schools’ and combinations within the community of Wittgenstein scholars. Wittgenstein himself clearly believed ordinary language to have an important role in philosophy, but different passages in his works from the same period can be interpreted to warrant views that, at least at surface, appear to be in conflict. In addition, there is the question of the role that ordinary language actually plays in Wittgenstein’s investigations and how to interpret that in the light of the explicit statements. Even with the best of intentions there can be a difference between what a person says and what he or she does in practice. Why could this also not be the case for Wittgenstein.

Hommen’s concern is an apparent inconsistency in Wittgenstein’s relation to ordinary language. On one side, Wittgenstein insists that ‘ordinary language is alright’ and, on the other side, he needs analogies and metaphors that are not part of ordinary language for his investigations and conclusions. The most famous is probably the account of language as akin to games – language games. Since he thereby goes beyond ordinary language, apparently ‘ordinary language is not alright’. Hence, we have a paradox.

While dissolving this inconsistency Hommen provides an inspiring account of what he calls Wittgenstein’s poetics (330) and its role in the perspicuous representations that the later Wittgenstein’s sought. The use of analogies and metaphors do not serve as effective summaries of key points but are also, or perhaps even more so, crucial components of the train of thoughts. According to Hommen, Wittgenstein’s analogies and metaphors are deployed to let us see language in a new light that dissolves the puzzles that had a grip on us, and secondly, shows us how the ambiguity of the meaning of words disappears the closer we get to use in real life. These points are important and often overlooked in the interpretation of Wittgenstein.

But Hommen gives it too much weight. Though important, the poetic approach cannot alone dissolve the puzzles that concerned Wittgenstein. Seeing new connections between previously unrelated familiar phenomena can both be illuminating and misleading. Wittgenstein distinguished between helpful and misleading analogies, and he did so not by measuring their aesthetic qualities, but through observation and the implications of those observations. Poetics is one in a set of methods that Wittgenstein applied, which to my knowledge still needs to be mapped out and linked. Ignoring other parts of the set, Hommen ends up attributing a view to Wittgenstein that makes him sympathetic to approaches he clearly opposed.

Furthermore, Hommen’s interpretation of ‘ordinary language’ is problematic and it forms an important underlying assumption for both the apparent inconsistency and the resolution of it. Although his interpretation might appear to respect what Wittgenstein wrote, at least on certain occasions, it cannot be reconciled with Wittgenstein’s practice, what he did. The interpretation which Hommen seems to adhere to, and which arguably can corroborate the apparent inconsistency that is the pivoting theme of the article, cannot be reconciled with the domains that concerned Wittgenstein, nor with long stretches of his argumentation. Adopting a different, prevalent interpretation of ‘ordinary language’ that accommodates some of these challenges and thereby arguably fits better to Wittgenstein’s use of the term also has the consequence that the initial inconsistency is removed. As Wittgenstein observes, the puzzles about languages do not occur in ordinary language under normal conditions but when we adopt a particular stance towards it. In that light, we cannot expect ordinary language under normal conditions to be suitable for the resolution of the puzzle.

For the reasons summarized above the argument of this paper is that the later Wittgenstein would not have held either of the two criteria that Hommen attributes to him to resolve the initial paradox. On the other hand, there is an interpretation of Wittgenstein’s approach to ordinary language available where the paradox doesn’t arise. Nonetheless, Hommen’s discussion of Wittgenstein’s poetics brings forward an important element in Wittgenstein’s argumentation and theoretical argumentation in general that to my knowledge is underexplored.

I begin with a recap of the position which Hommen attributes to the later Wittgenstein and the underlying motivation for the two criteria which Hommen believes Wittgenstein would or should have used to decide on which philosophical accounts to be sympathetic to. The next three sections will discuss the semantic criterion, and the two sections hereafter the epistemic criterion. Each of the discussions will bring forward elements of Wittgenstein’s practice that conflict with the criteria, even if his writings could be interpreted to warrant attributing them to him.

2 Wittgenstein’s Position according to Hommen

Wittgenstein repeatedly criticizes the advancement of philosophical theories and theorizing. As such it makes sense that he should only concern himself with phenomena that lie open for anyone to grasp, such as ordinary language, and refrain from moving beyond description in his philosophical endeavors. The question, however, is whether Wittgenstein nonetheless ends up falling prey to his own criticism.

As Hommen says, there appears to be a certain inconsistency in the later Wittgenstein’s approach to ordinary language. According to the later Wittgenstein, the philosopher,

1. “… may in no way interfere with the actual use of language” (PI, §124) and “… leaves everything as it is” (ibid.), and

2. “… seeks philosophical redemption in the ‘language of every day’ (PI, § 120) and consequently adheres to a pointedly casual, colloquial style in his own lectures and writings: …” (315).

While on the other hand, he holds that to attain,

1. “… a perspicuous representation of the grammar of ordinary language, one should find – and even invent – intermediate cases (whatever these are)” (322), and

2. “… many of the terms Wittgenstein coins in his later writings seem to be highly technical. … Wittgenstein uses them in such peculiar, constricted or extended (if not clearly defined) ways … (cf. Read 2005, p. 83).” (315)

From a deductive logical point of view the observed inconsistency appears to be weak, since it is possible to hold that, i) the object of study is not to be interfered although thought experiments are designed for the purpose of analysis, and ii) summarizations and illustrations of findings can introduce new concepts and cases, even though most of the time the redemption is sought in everyday language. But there is certainly an apparent tension, because as mentioned in the introduction, on one side, Wittgenstein insists that ‘ordinary language is alright’ and, on the other side, he needs analogies and metaphors that are not part of ordinary language for his investigations and conclusions, and hence it would appear that ‘ordinary language is not alright’.

Hommen probes the question whether these contravening tendencies might be reconciled (315). His conclusion is that the abovementioned inconsistencies are only appearances, because using technical terms that are part of a conceptual system that dispels confusion in ordinary language can be admitted if these are also ‘semantically accessible to ordinary language speakers’:

“For Wittgenstein, there are, in the end, no philosophical criteria for choosing particular conceptual systems except two: first, any candidate system must be semantically accessible to ordinary language speakers; second, any such system should enable an overview of the phenomena under consideration and help to dispel confusions that result from straitened perspectives on those phenomena.” (332)

Following these criteria, we can arrive at a perspicuous representation of the grammar of ordinary language (322) that Wittgenstein describes as of fundamental significance (PI, §122). It would seem then that the later Wittgenstein was a pluralist or perhaps even a relativist in the pursuit of the resolution of philosophical puzzles and paradoxes because these criteria could in principle be met in several ways.

In this way, then, Hommen can resolve the apparent paradox in the later Wittgenstein’s approach. The ordinary language of common sense is the bedrock against which any language of philosophical inquiry must be compared (320). But to see ordinary language in the proper light and comprehend connections hidden to us, metaphors and analogies are powerful methods which are applied and accepted by the later Wittgenstein because they create new connections between domains and in that way contribute to ‘dispel confusions that result from straitened perspectives.’

I will call the first of Hommen’s criteria the semantic criterion:

The Semantic Criterion: A particular conceptual system can be chosen to provide a perspicuous representation of the grammar of ordinary language, if the system is semantically accessible to ordinary language speakers.

The second criterion I will refer to as the epistemic criterion:

The Epistemic Criterion: A particular conceptual system can be chosen to provide a perspicuous representation of the grammar of ordinary language, if the system enables an overview of the phenomena under consideration and helps to dispel confusions that result from straitened perspectives on those phenomena.

To get a better understanding of their respective contribution to lifting the contravening tendencies, we will examine the argumentation for each of the criteria separately beginning with the semantic criterion.

3 The Semantic Criterion

The first criterion stresses the primacy of ordinary language and accommodates Wittgenstein’s emphasis on the sanctity of ordinary language. But what is ordinary language? Hommen does not explicitly define his understanding of this critical term, but characterizes it on several occasions:

1. “Wittgenstein seeks philosophical redemption in the “language of every day” (PI, § 120).” (315)

2. “According to Wittgenstein, ordinary language is a direct expression of human nature. Reflecting our primordial needs and interests, it is what channels our experiences and guides our actions, determining, as it were, our basic interpretation of the world as well as our peculiar way of coping with it. As such, ordinary language “pervades all our life” (BB, p. 59)” (317)

3. “… what applies to languages in general applies to professional jargons and technical terminologies in particular: they must always be (re)translatable to the vernacular of ordinary speakers in order to be intelligible at all. Thus, what can ever be said about anything (our experiences, life, the world), must be sayable – in principle, at least, if not in practice – in everyday language.” (318–19)

4. “… nomenclatures and notations may sometimes be useful for clarifying or simplifying our customary concepts and distinctions. Such improvements, however, can never exceed the resources of ordinary language. After all, they have to be acknowledged by us – and that means that we have to carry out the clarifying, simplifying, or whatever, in our common idiom (cf. Hanfling 2000, p. 160).” (319)

5. “… the one language independent of all the others and therefore the standard against which any candidate language of philosophical inquiry is to be measured is the ordinary language of common sense. The latter is, as Wittgenstein says, the “hard bedrock, deeper than any special methods and language-games” (RPP I, § 648), where we eventually reach the “facts of living” (RPP I, § 630): the ineluctable point of departure – and point of arrival – for all our endeavors.” (320–21)

It seems reasonable to interpret Hommen’s view as suggesting that by ‘ordinary language’, Wittgenstein referred to the vernacular of everyday speakers – everyday language – that possesses a particular authoritative relation to reality as it reflects our primordial needs and determines our basic interpretation of the world from where any deeper endeavors depart. It is the independent foundation of our relation to the world, and other conceptual systems and technical terminology must, in practice or principle, be able to be sayable in everyday language.

Accordingly, the later Wittgenstein held that problems in philosophy arise from misinterpretations of surface grammar (PI, § 664) and other distortions of ordinary language. In fact, ‘puzzle’ would seem to be a more appropriate term than ‘problem’ because it is not solved by acquiring new information, but rather by realizing how the pieces in front of you belong together (327). Given the relation between ordinary and technical language, the former being our point of departure, it would seem to imply that the puzzles arising in everyday language have a deeper quality than potential puzzles in a technical language, because they concern the direct relation between human and world whereas a puzzle in a non-foundational branch of language only indirectly does so.

The solution of a puzzle in philosophy must also be given in ordinary language and be understandable to those who experience it – of which we can only assume that they are ordinary language speakers. Wittgenstein not only lends primacy to puzzles arising in ordinary language. The explanation or solution to the puzzle must be “… sayable – in principle, at least, if not in practice – in everyday language. There can simply be no insights, philosophical or otherwise, categorically barred from the understanding of common sense; for what is barred from our apprehension, can be no insight.” (319). In fact, ‘dissolve’ would seem to be a more appropriate term than ‘solve’ because insight is not achieved by arranging the pieces in a new way, but by realizing that it was a misinterpretation that started the train of thought that the pieces needed to be rearranged.

A consequence of the framework laid out above in the interpretation of ordinary language is that a revisionist approach to ordinary language would be difficult to formulate. This is so, because dissolving a philosophical puzzle of everyday language by saying that we should speak in a different way only can be made understood by relating it to ordinary language which in essence creates a full circle back to ordinary language.^[5] If we find ourselves somehow moving outside of everyday language, we need to make sure we have a clear route to getting back because our questions and our answers are at risk of making sense. Hommen’s Wittgenstein interprets ordinary language, understood as the vernacular of everyday language, as the starting point and ending point of his investigations: That which needs to be explained (explanandum) and that which contains the explanation (explanans) must be given in, or translatable into, ordinary language, and thereby ‘semantically accessible’ to ordinary language speakers.

How, then, to make progress in our understanding if we, unlike scientists, are not allowed to introduce new theoretical entities or causal relations? Given the above constraints we are still allowed to look at how language is used. We can point out where generalizations and other assumptions are unwarranted in the sense of not being an actual description of how we use language. And we can point to previously unseen similarities between how words are used and language functions in domains that were previously seen as unrelated. Thereby we can hopefully dismantle the confusion about how we as humans relate to the world: “The real discovery is the one that enables me to break off philosophizing when I want to.” (PI, §133).

4 Wittgenstein’s Practice and the Semantic Criterion

The issue with the above provided account of the description and role of ordinary language in the later Wittgenstein’s works is that, although several of his remarks could be interpreted to warrant this view, it is very difficult to align with the topics that concerned Wittgenstein or with his argumentative approach.

To begin, if Wittgenstein adhered to the semantic criterion, that could perhaps make sense of the fact that commonly used terms such as ‘meaning,’ ‘knowledge,’ and ‘pain’ are among the most discussed in the Philosophical Investigations. However, it does not account for the fact that over half of Wittgenstein’s writings from 1929 to 1944 are devoted to mathematics.^[6] For instance, the puzzles concerning infinity that arise in relation to diagonal proof in set theory or Gödel’s incompleteness theorems – areas in which Wittgenstein invested considerable time – can hardly be considered part of everyday language, except perhaps for a small group of mathematicians. In-principle-‘sayability’ might be invoked to say that problems in mathematical philosophy can be stated in everyday language, or perhaps is a variant of a puzzle that also exists in everyday languages. Using the phrase ‘in principle’, probably several times, the argumentation would have two steps: The first step would be to argue that the mathematical language puzzle could be translated into an everyday language puzzle. The next step would be to dissolve the everyday puzzle and by inference the mathematical language puzzle. But Wittgenstein never took this approach and appealed to ‘everyday language’ at critical parts of his investigations into mathematics.

Secondly, the emphasis on ordinary language as foundational in the argumentation creates a framework for acceptable explanations that does not align with Wittgenstein’s approach, as it leaves out considerable parts of his investigations and argumentation. If ordinary language is foundational, then it should suffice as an argument against a position that it does not respect the practices followed by ordinary language. A theory of pain, for example, whether it posits private entities or specific types of behavior as truth conditions, can be refuted within this framework by pointing out that it does not honor the way we talk about pain: The philosopher, by drawing hasty and skewed conclusions about our concepts of pain, has been misled into a false understanding of what pain is and what he or she is supposed to justify.

An example here is the famous passage about private language usually taken to being at §243 in the Philosophical Investigations, where Wittgenstein introduces the question of whether an in principle private language is possible: “The words of this language are to refer to what only the speaker can know – to his immediate private sensations. So another person cannot understand the language.” (PI, §243). In the paragraphs that follow, he describes how our ordinary sensation language does not resemble such a language by showing that in ordinary sensation language we can in most cases not only understand each other but also know when another person is in pain. When we say things like ‘sensations are private’ we do not talk about privileged knowledge or differences between individuals in understanding. Instead, it is most often used to explain that part of the concept of sensation involves a difference between having a sensation and looking at someone having a sensation. It is analogous to saying, ‘one plays patience by oneself’ (PI, §248) and ‘every rod has a length’ (PI, §251). In §256, he sums up and returns to the original question:

Now, what about the language which describes my inner experiences and which only I myself can understand? How do I use words to signify my sensations? - As we ordinarily do? Then are my words for sensations tied up with my natural expressions of sensation? In that case my language is not a ‘private’ one. Someone else might understand it as well as I. – But suppose I didn’t have any natural expression of sensation, but only had sensations? (PI, §256)

From here the focus shifts to a more principled discussion about the possibility of a private language. The argumentation is not easy to follow and its course is still subject to controversy, but it is clear among his conclusions, that if ‘pain’ referred to a privately held object for each of us, meaning that each of us would be the sole judge of whether we were in pain and any so-called pain behavior would only be an indicator, this object would be irrelevant to our ordinary use of the word ‘pain’. He considers several ways in how the private object could be presented to the person having the experience and ends up with the same conclusion.

Up until §256, Wittgenstein shows that in everyday language we do not speak as if we can’t know or understand the sensations of another person and introduces analogies to make sense of a sentence like ‘sensations are private’. But he doesn’t stop there: It is not enough for a perspicuous representation that ‘enables him to break off philosophizing when he wants to’.

Hommen’s semantic criterion ultimately brings the later Wittgenstein’s views on language close to a fundamentalist approach with its emphasis on everyday language possessing a particular authoritative relation to reality. There seems to be warrant for the criterion in what the later Wittgenstein writes about ordinary language and its role in philosophical puzzles and dissolutions of them. But the criterion goes against what Wittgenstein does: The topics that concerned him and argumentation that solves the problem.

5 Context Situating

What would be an alternative to interpreting Wittgenstein use of the term ‘ordinary language’ as meaning the vernacular everyday language which could accommodate the variety of topics that concerned him and the breadth of argumentation methods? It is not within scope of this article to present a fully-fledged alternative, but there is a path if one allows the context in which concepts are used to play a more prominent role. Wittgenstein’s repeated references to ordinary language could be interpreted to refer, more accurately, to the ordinary use of language in a context as hard bedrock, regardless of whether that context is mathematical logic discourse or describing sensations of pain in a doctor’s office. There would be a few important differences in such an approach compared to Hommen’s that are worth pointing out.

There is ample room for context to play an important role in Hommen’s account of Wittgenstein’s view on the role of ordinary language. Within this account, it is still possible to hold that sentences, words, and entire linguistic practices must be understood in the context in which they are used. ‘Context’ here understood broadly to mean any circumstances that form the setting for an event, statement, or idea, and in terms of which it can be fully understood. However, as everyday language stands in this basic relation to the world, statements in all other contexts must connect to this special context or context-free discourse. This is the requirement of semantic accessibility for any ordinary language speaker. A use theory of meaning which leaves out this requirement, could see all contexts in which statements are uttered as hard bedrock for the explanation of the statement’s meaning understood as use. According to such an approach, if concepts and words have found applicability in a context, then there is really no need to require that it should be somehow related to the vernacular, everyday language, whatever that is, for it to be meaningful. There is little more to say than that the meanings of the words can be derived from their use, but also that meanings can, in many cases, be context-specific because the same word in a different context can have a similar yet distinct usage. There is nothing sacrosanct about the vernacular in the sense of its standing in a special relationship to reality or ‘form of life’: Languages or words that have found use in a context cannot be argued to not function, exactly because they have a use in this context. The actual use in a context is an equilibrium reached through a historical process. This use can certainly evolve and there is nothing sacrosanct about the language we use at a certain point in time.^[7] Accordingly, there is such a thing as the ordinary language of set theory, namely the language that proficient mathematicians use when conversing in this branch of mathematics. As with any language or branch of language, set theory discourse can generate philosophical puzzles that Wittgenstein could find intriguing, perhaps disturbing is a better term, enough to allocate tie for investigating.

A potentially felt weakness of a context-situating approach to language as opposed to a vernacular-basic approach, could be that the former will emphasize context as bedrock as opposed to semantic accessibility to everyday speakers This makes it somewhat difficult to draw general conclusions about actual use. Minimal changes in context, such as who you are talking to, can change the use of terms and thereby the meaning. How, then, to say something that dissolves a general puzzle about the use of ‘pain’ if every context in which ‘pain’ figures has its own use? For instance, contexts might vary in how much they allow first-person reports of pain. If someone is badly hurt from a car accident but claims not to feel any pain, the paramedic would probably use that as a truthful statement indicating a serious fracture affecting nerves. On the other hand, someone with their eyes full of tears during a relationship break-up might not be trusted if claiming ‘I feel no pain’. How, then, to draw any conclusions that are universally valid or even just generalizable?

This consequence might be seen as a weakness of a context-situating approach, but it aligns well with Wittgenstein’s criticism of philosophers’ ambitions to solve universal issues once and for all:

It is as if we wanted to grasp the unlimited strips and complained that it can’t be done piecemeal. Of course, it can’t, if by a piece one means an infinite longitudinal strip. But it may well be done, if one means a cross-strip. – But in that case we never get to the end of our work! – Of course not, for it has no end.

We want to replace wild conjectures and explanations by quiet weighing of linguistic facts.^[8]

In this passage, Wittgenstein essentially describes a shift in the focus of philosophy from solving universal issues once and for all to a continuous, infinite dissolution of misinterpretations of language. Also, this consequence of a piecemeal approach fares well Wittgenstein’s continued hammering at a puzzle, continued rephrasing of statements of the puzzle, and argumentation not only from actual practice, and analogy, but also from a logical point of view, trying to extract some general lessons and warning signs that could short cuts or general approaches to dissolving issues in other contexts. Interpreting Wittgenstein’s ‘ordinary language’ to refer to something along the lines of ‘ordinary use of language in a context’ in this way indicates a way to accommodate and respect the way Wittgenstein worked – what he did: The topics that concerned him and the breadth of argumentation methods.

There is another aspect of Wittgenstein’s work that also doesn’t fit well an account which requires the perspicuous representation, the dissolution of a puzzle, to be ‘semantically accessible to ordinary language speakers.’ According to Wittgenstein, as quoted by Hommen, ordinary language is of interest because of its role in setting up traps: “When we do philosophy we are like savages, primitive people, who hear the expressions of civilized men, put a false interpretation on them, and then draw the queerest conclusions from it. (PI, § 194)” (321). This, however, does not reconcile well with another famous statement that ‘ordinary language is alright’. Anything which again and again leads to false interpretation can hardly be said to be alright.

Obviously, we are here confusing two ways of approaching ‘ordinary language’ or ‘the expressions of civilized men’. Ordinary language speakers, or civilized men, use language in all kinds of contexts in which language is alright in the sense that it serves a function and is reproduced. Philosophers, on the other hand, adopt a different approach. They observe language, isolate expressions from their context and interpret them outside of use. The effect is that the philosopher’s take on language is no longer as ordinary language speaker. During life, the problems or puzzles related to everyday terms such as meaning, knowledge, pain and so forth that concerned Wittgenstein do not arise in ordinary use of everyday language. Even when you feel that you have grasped a problem, it can be very difficult to explain to proficient ordinary language speakers and convince them that there is an interesting problem here. It is not enough for a proficient user of language to be introduced to the problem to come under its spell; he needs to leave his or her natural stance. Hommen via the semantic criterion requires the dissolution to be understandable by ordinary language speakers, but ordinary language speakers can’t even understand the problem. If you understand the problem, that means you have moved away from the position held by ordinary language speakers. What Wittgenstein essentially offers is a way back: “What is your aim in philosophy? To show the fly the way out of the fly-bottle.” (PI, §309)

6 The Epistemic Criterion

The second criterion is called the epistemic criterion, because achieving overview and dispelling confusion are essentially epistemic notions. The resolution of philosophical puzzles resides in providing a perspicuous representation and for this purpose intermediate cases such as the primitive language in the beginning of Philosophical Investigations can be helpful irrespective of whether they exist or not:

As he elaborates, the required links are not, or need not be, causes – indeed, it does not even matter whether they actually exist or not. The intermediate cases merely serve as (imaginary) objects of comparison whose purpose is to create fruitful analogies that can unify the phenomena in the domain of observation (cf. Baker and Hacker 2005, p. 312). (324)

Indeed, it is a tool in Wittgenstein’s poetics that can be helpful to free our minds from ‘pictures’ that hold us captive (PI, §115). Another one is the use of analogies:

More specifically, Wittgenstein’s poetics is characterized by the iterative and recursive deployment of analogical and metaphorical figures. Consider, for example, Wittgenstein’s comparison of words with chess pieces (cf. PI, § 108), his analogy of languages and games (cf. PI, § 7), and the association of these with the human “form of life” (PI, § 19) which is the “river-bed of thoughts” (OC, § 97) and the “bedrock” (PI, § 217) of justification: the recurring juxtaposition of these and related similes – all taken from ordinary language, so as to “let our thoughts roam up and down in the familiar surroundings of the words” (Z, § 155) – creates a layering of meanings which tracks down and condenses several, seemingly inconsistent aspects of linguistic meaning: (330)

To Hommen, then, Wittgenstein’s elucidations are achieved by use of a particular conceptual framework, a philosophical poetics (330).

This approach is a universal one. Hommen puts forth several points regarding the connection to scientific theories and how Wittgenstein himself contemplated the role of poetic language to inspire as part of their persuasive potential. Persuasion is an epistemic dimension, and a position’s persuasiveness lies not only in how it describes phenomena but also in its aesthetic potential to create a coherent narrative and analogies and inspire further inquiry. This is similarly applicable in philosophy and played a role in Wittgenstein’s reflections on his own method. The aesthetic experience, when after, probably repeated, readings of passages in Philosophical Investigations one sees the analogy between language and games, shares similarities with the aesthetic experience one has when reading and gradually building a better understanding of important theories such as those of Copernicus, Darwin, and Freud, or learning a new theoretical formula in mathematics or logic. Perhaps it is even indispensable for “facilitating a comprehension of contents which would otherwise be incomprehensible?” (331). The feature of introducing a new way of thinking or looking at a section of the world is shared between Wittgenstein’s approach and paradigm-shifting scientific theories.

The effect of being presented with an intermediate case, analogy or a new scientific framework can be described by the phenomenon of ‘seeing-as’ or ‘aspect seeing’ to which Wittgenstein devoted considerable time. The duck-rabbit illusion is a famous example, but also when a person realizes a similarity between two faces: “I contemplate a face, and then suddenly notice its likeness to another, I see that it has not changed, and yet I see it differently.” (PI, II, 193).

Aspect-perception is strongly reminiscent of the state of ‘seeing the connections’ which is reached in a perspicuous representation. As in the latter case, what is perceived in the dawning of an aspect is not a newly discovered property of the percept – a further fact – but rather “an internal relation between it and other objects” (PI II, p. 212). And as in the case of ‘finding connecting links,’ Wittgenstein emphasizes the creative moment in aspect-perception: “The concept of an aspect is akin to the concept of an image. […] Doesn’t it take imagination to hear something as a variation on a particular theme?” (ibid., p. 213). (325)

According to Hommen, Wittgenstein sees the dawning of a new aspect is essentially an aesthetic experience that unveils internal relations between objects. The true philosopher should have a strong aesthetic sense:

In a telling summary of his philosophical method, Wittgenstein states: “really one should write philosophy only as one writes a poem” (CV, p. 28). This confession highlights the perhaps most important facet of Wittgenstein’s aesthetic view of philosophy. The peculiar knowledge that consists in seeing the connections and in the noticing of aspects is, in the end, ineffable. (328)

Following this poetic ambition, according to Hommen, Wittgenstein’s analogies and metaphors are deployed to let us see language and its puzzles in a new light that dissolves the puzzles that had a grip on us, and secondly, shows us how the ambiguity of the meaning of words disappears the closer we get to use in real life.

Hommen concludes that Wittgenstein’s use of analogies and intermediate cases not only brings his approach to philosophical puzzles closer to aesthetics, but also that Wittgenstein must be quite liberal regarding which philosophical languages may assist us in our philosophical musings:

“many different styles and façons de parler may prove to be convenient for the respective purposes of our different philosophical endeavors – be it the ordinary language of everyday life, the formal framework of the logician, or the transcendental jargon of the metaphysician.” (332).

Wittgenstein’s aesthetic conception of philosophy allows for poetic conceptualization and combination (332) and is essentially pluralist with regards to the analogies or meanings of a sentence that carry the elucidation.

7 The Epistemic Criterion and Wittgenstein’s Practice

The later Wittgenstein has historically been situated within traditions that have placed less emphasis on aesthetic and poetic elements to generate insight and clarity.^[9] As part of the Oxford school or ordinary language philosophy, there has been focus on his detailed analyses of how language is actually used as method to dismantle philosophical problems.^[10] However, since the early 1970s his influence might have been even greater due to the arguments presented in Philosophical Investigations, that derive controversial conclusions from apparently trivial assumptions: The rule-following considerations, which became an independent topic in the philosophy of language following Saul Kripke’s interpretation,^[11] and the so-called private language argument, which has been the most discussed topic in the secondary literature.^[12]

In this tradition, Wittgenstein’s poetic approach and thereby his use of analogies and metaphors serve as effective summaries of key points, but there has been less focus on the potential of poetic language itself to persuade or provide transparency – to carry the argumentation. Hommen argues that Wittgenstein’s aesthetics play an important role in his approach.^[13] Analogies do not just summarize but are the essential component that dispels or removes confusion.

But, as shown above while discussing the paragraphs following §256 in Philosophical Investigations, there are several elements in Wittgenstein’s investigations that carry the argument. Poetics is only one element. The argument in this section will be that the perceived clarity of metaphors and conceptual systems – their potential to dispel confusion – is not the standard by which Wittgenstein judges their validity. It is judged their potential to summarize or withstand more detailed dissection.

Wittgenstein uses analogies, images, and intermediate cases both to support conclusions he defends and the views of his opponent, so without a ballast to interpret these, a poetics will point towards both groups of views. His analogy between language and games is introduced very early in philosophical studies in connection with a critique of Augustine’s picture of the nature of human language:

It is as if someone were to say, “Playing a game consists in moving objects about on a surface according to certain rules.” and we replied: You seem to be thinking of board-games, but they are not all the games there are. You can rectify your explanation by expressly restricting it to those games. (PI, §2)

It is this comparison that was later condensed into the concept of ‘language games’ which, for better or worse, has been a metaphor that has guided the understanding of the late Wittgenstein. But the early Wittgenstein’s comparison of sentences as images of reality can, based on its prevalence, be said to be an equally powerful metaphor that for a long period set, and still sets in significant discussions in analytical philosophy of language, the framework for resolving the confusion about how to understand language as a phenomenon:

2.151 The form of representation is the possibility that the things are combined with one another as are the elements of the picture. 2.1511 Thus the picture is linked with reality; it reaches up to it. 2.1512 It is like a scale applied to reality. (TLP)

The later Wittgenstein rejected that image, but hardly based on its lack of aesthetic qualities. It was because it was not a correct description of the nature of language. It collapsed when he tried to unfold it.

If one had only metaphors, then even within the framework of Philosophical investigations there are also various metaphors that without the surrounding descriptive context can be difficult to assess Wittgenstein’s stance on. Whether he advocated them or used them to describe misguided views. Compare, for example, these:

a cry, an expression of a pain – a sentence, an expression of a thought. (PI, §317)

The sentence “Sensations are private” is comparable to “One plays patience by oneself”. (PI, §248)

In aesthetics, a playful attitude is adopted e.g. towards language and the possibilities for people, including oneself, to be affected, moved, and given new inspiration is explored. I can only speak for myself when I claim that both statements connect two separate domains in a way that new ideas are generated and there is a hint of clarity that inspires to be a basic idea that can be built on. But it is not those parameters that ultimately define their valuation or usefulness. Wittgenstein concretizes and examines the analogy in depth, and it is the summary of that analysis that ultimately determines its durability. The philosophical qualities of the analogy are not only constituted by the aesthetic qualities.

The requirement of withstanding a depth analysis does not as imply that Wittgenstein could not be a pluralist with regards to which analogies and intermediate cases can be most fruitfully applied in the description of different areas of language. There is scope for different styles and façons de parler in the selection of metaphors and in that sense Wittgenstein is a pluralist. At that level, it cannot be ruled out that formal languages ​​​​or metaphysical concepts may be appropriate. But it should come with a warning label:

When philosophers use a word a “knowledge”, “being”, “object”, “I”, “proposition/sentence”, “name” - and try to grasp the essence of the thing, one must always ask oneself: is the word ever actually used in this way in the language in which it is at home? – 

What we do is to bring words back from their metaphysical to their everyday use. (PI, §116).

The risk of becoming misunderstood or overinterpreted if one introduces formal or metaphysical terms even merely as analogies should be obvious. On occasion it seems that Wittgenstein was willing to take that risk. See for example “Essence is expressed in grammar.”^[14]

Wittgenstein himself had been captivated by the picture (PI, §115) that a perspicuous representation of the grammar of ordinary language would requires a system that could facilitate a logical breakdown to elementary propositions. Driven by its own metaphor of an underlying layer of transparency and grounding that dispels confusion by reconstruction, the idea is to create a translation which allows us to see disagreement and puzzles in a proper light. In the late 1920s when he returned to philosophy after a long absence, this fundamental layer of meaning had to be found in sensory experiences described in what he called a ‘phenomenological language’. His only published article was written in this spirit,^[15] and his diary entries clearly reflect his desperate pursuit of the idea of a phenomenological language that he ultimately realizes must be abandoned.^[16]

By the early 1930s Wittgenstein had rejected this type of approach yet found it compelling enough to continually return to it as a counter position. He had tested the analogy to its limits and found it utterly misleading. There is no formal language whose rules are completely clear to which we can translate ordinary language and thereby dispel confusion. That analogy is an illusion, and such an application of formal languages would not be consistent with later Wittgenstein’s understanding of semantics.

Hommen’s second criterion cannot stand alone as an epistemic requirement. It is not enough for a conceptual system to seek to create transparency by alleviating confusion. The criterion itself cannot distinguish between an approach that Wittgenstein can be assumed to be sympathetic towards and approaches that seek this transparency through a linkage to a fundamental level of meaning. The latter being an approach that Wittgenstein, at one time, sought inspiration from, but ultimately rejected.

8 Concluding Remarks

Hommen presents several thought-provoking points regarding the later Wittgenstein’s reflections on the persuasive power of his arguments and how philosophy and poetry have things in common. He reaches this conclusion through an inspiring and persuasive account of Wittgenstein’s use of analogy and metaphors. It is demonstrated how the later Wittgenstein likens his approach to a philosophical problem or puzzle to how a poet perceives connections and notices details (328). Wittgenstein himself had multiple reflections on this similarity, which he regarded as more than merely superficial: “Really one should write philosophy only as one writes a poem” (CV, p. 28). By employing analogy and metaphor, we can “free our minds from ‘pictures’ that hold us ‘captive’ (PI, § 115),” that is, from descriptions of the phenomena of language that are too narrow or one-sided, which can lead us into philosophical disasters.

The points are made, however, while presenting a view of the later Wittgenstein’s approach to philosophy that attributes to him views on ordinary language and what it requires to dispel confusion that it is difficult to believe he would have held when looking at how he practiced philosophy. Even though some of his statements seem to indicate otherwise.