It’s difficult to go from philosopher to philosopher and try to get an idea of what it is they are trying to do in general. After getting into philosophers you think get to the core of epistemological problems, to delve back into strange metalanguage studies is quite tiresome. Michael Dummett does not seem to be a straight forward thinker. The Seas of Language prefaces itself with the fact that the views expressed in several of the essays contained within are no longer accepted by the writer himself. That makes problematic the attempt to really get into a thinker with the idea of any consistency or ground that drives the thought of the thinker. Instead, it seems, the analysis will have to consist of attacking or bolstering specific lines, without necessarily relating any of it back to a system of thought.

Even in disparate or non-coherent essays you can find some key thoughts or sentences that spark something, or can be worth considering. Even in the worst of thinkers there can be points of salience, there can be accidental genius, great expressions, and so on. Despite not wanting to, it might be good to read through the torture, if only to better know what makes the experience so miserable.

Dummett does have his moments: “To give an adequate explanation of the capacity in question [the capacity for knowledge], the account must do more than simply specify the fact that must be known: it must indicate how, in particular, awareness of that fact must have been attained, that is, what process of derivation is required for it to count as knowledge, in the strict sense.” (p. 10) Of course, this is something I can agree with and something many philosophers try to respond to. What is the causal progress towards knowledge, and what, more than awareness, is required for it to be counted as such? It is not enough to simply become-aware of something, that is not communicable or interesting on its own. It seems Dummett have a similar idea, and that it is closely linked to language-use. To know is not simply to be aware of or apprehend mentally (whatever this could mean), but it is to talk about and talk about it correctly. A lot of Dummett starts from a point of view from the Wittgensteinian behavior approach to language. You know a language when you know how to use it. But this does not solve the notion of meaning. What does each component of a language mean, and how is it determined? When Dummett speaks of metalanguages it really does nothing to get closer to an answer to this question. To understand the truth of something like “’The earth’ denotes the earth” is entirely tautological and pointless to say. It does absolutely nothing. Things like this are what make philosophers of the analytical tradition quite tedious at times, since it seems they create a level of problems that are no longer connected with the core problematic. Although this might not be an analytical problem, just a problem with confused philosophy.

So when Dummet writes “anyone who knows the use of ‘denotes’, and who knows that ‘the Earth’ is a singular term in English, must know that the sentence “’The Earth’ denotes the Earth” is true, even if he does not know what, specifically, the phrase “the Earth” means or what it denotes.” (p. 12) he does nothing but show that people can understand a sentence that has nothing to do with knowledge or meaning proper. It’s on par with knowing that 1+1 = 2, it has no real value for any type of knowledge theory other than to make a distinction between truths of the mental kind, thoughts corresponding to each other logically, and meaning as it is created concretely in the world as relation to a world. But you have to also ask yourself if it’s even true. How can you know the truth of a term if you do not know what it means? Does just saying “The Earth” denote anything? It seems to me that a phrase does not necessarily have to be intended as directionality towards an object. How can we know how to use the Earth if we do not know what it means? There is something about the above-mentioned sentence that just doesn’t work. It seems devoid of concretion, and very far removed from the world of actual language-use.

“One can say only that the knowledge of the entire theory of truth issues in an ability to speak the language, and, in particular, in a propensity to recognize sentences of it as true under conditions corresponding, by and large, with those stated by the T-sentences.” (p. 16) But again, what exactly does it mean to be able to speak a language, and what does it mean for truth-sentences to correspond to particular conditions?

His first essay “What is a theory of Meaning?” does not in fact go into the notion of what that would be. But seem to be a preliminary dealing with various forms of meaning and the criteria which would then later, once you’ve been guided by these criteria, lead to an idea of meaning. Holism cannot be it, since it does not sufficiently ground meaning. Holistic semantic views seem very difficult to understand in any concrete sense, to me. I am not quite sure what it means to talk about holism. A one-word sentence surely carries meaning on its own, to some extent, if only to notify somebody of the existence of some thing in their field of experience. We explain meaning through reference to or with further language-use, but if we keep testing the meaning of what we are trying to get at, we are eventually coming down to very basic observation sentences, and the notion of denotation, where what is denoted is no longer language, but some kind of truth-condition, or condition, state of affairs, experience, observation, etc. Surely this is where meaning is formed as its most primitive and fundamental core, as the atoms that then give meaning to larger attempts at language-use. This progression is how we acquire language, and thus the meaning (if meaning is language-use, that is, proper language-use) of speech acts. Dummett seems to hold a similar view, based on his concluding words in the essay, but I am not at all sure about what lead him to this point, although there are phrases I can get behind (p. 21):

I am not objecting to the idea of a theoretical representation of a practical ability as such, and certainly not to the representation of a mastery of language by means of a deductive theory: I am saying only that such a representation is devoid of explanatory power unless a grasp of the individual propositions of the theory is explained in terms of a specific practical capacity of the speaker. I do not know whether this is possible; I do not know that holism is an incorrect conception of language. But I am asserting that the acceptance of holism should lead to the conclusion that any systematic theory of meaning is impossible, and that the attempt to resist this conclusion can lead only to the construction of pseudo-theories; my own preference is, therefore, to assume as a methodological principle that holism is false.

I would agree with this by virtue of what was said before the quote. But even more important is what precedes it, which also informs the quote above:

In many contexts, we may take as unproblematic the ascription to someone of awareness of some fact, since we may credit him with an understanding of language, and the manifestation of his awareness will consist primarily in his ability to state the fact or his propensity to assent to a statement of it. But, where we are concerned with a representation in terms of propositional knowledge of some practical ability, and, in particular, where that practical ability is precisely the mastery of a language, it is incumbent upon us, if our account is to be explanatory, not only to specify what someone has to know for him to have that ability, but also what it is for him to have that knowledge, that is, what we are taking as constituting a manifestation of a knowledge of those propositions; if we fail to do this, then the connection will not be made between the theoretical representation and the practical ability it is intended to represent.

So to say you have knowledge and proper language-use, you have to have some kind of theoretical approach on how this could come about. There cannot simply be coherence without any way to determine coherence, obviously. Understanding is not clear whatsoever, and it does not come about through assent alone, I don’t think, despite the partial agreement I have with Quine’s notion of it. Because one always have to ask what informs the assent or dissent, and I am not too sure I’d be happy to leave this answered by intuition alone, since there are conscious ways of explicating this process, first-hand, even. Again, recourse to Husserl would be suitable. One context of language-use is understood, despite being removed from other contexts where similar or the same words are being used, in my view. And this is a core principle, because it makes language possible to ground somewhat outside of itself, in a way that incorporates experience, apperception and thus intentionality. We clearly often forget past language-use, and that is not always what grounds our use of language. Although it might be included in how we see a language’s meaning in general, as a collection of different times of use and their respective contexts, in each moment we do not recall each such past moment. There is something much quicker in place, a habitual speed, a much more accessible thing, which informs our use. It is not as such the past contexts that inform our use, it is whatever remains of these past usages that can be carried with us in everyday language-use that makes expression and meaning possible. Dummett makes similar comments on page 31, the appendix to the first essay “What is a theory of Meaning?” on the second paragraph. It might be worth quoting:

If a speaker’s mastery of his language consists in an implicit grasp of a theory of meaning for that language, then, if the theory is holistic, he must be aware of the judgements which comprise the base totality. Even when the language is his own personal idiolect, therefore, that totality cannot contain a multitude of casual judgements which he has made but has subsequently forgotten; it can, at any given time, contain only such judgements as can be elicited from him at that time. This still makes it grossly improbable that the totality can be sufficiently extensive to determine the references of all the words in his language.

But of course I am fucking flailing helplessly trying to understand this, and it would be a lot nicer if I was not a complete idiot with no ability to properly focus and retain philosophical concepts in a way that is necessary to even remotely have worthwhile things to say regarding these topics.

Dummett comes closer to an idea of meaning that is constructive in his second essay, “What is a Theory of Meaning? (II)”. Here meaning are things like inductive specification, which is the same as reference, it seems. “The meaning of a sentence is the method of its verification” is another way Dummett puts it. So in this essay Dummett decides to fall completely into the systematic approach, where meaning is more clearly defined. That the meaning of a sentence is the method of its verification means it is possible to create a rigid definition of specific meaning in specific sentences. To say “this is a tree” can be verified by pointing to a tree. Or “this is what we call a tree”, would perhaps be more suitable example. The truth-condition is then the ostensive act coupled with the speech act, and so on. He simply wants to denote by this kind of view that there is a commonality to the ways we use language, and perhaps even, although not quite expressly said, that there is a structure to the way we form sentences in relation to various conditions: “there is some uniform means of deriving all the other features of the use of any sentence from this one feature, so that knowledge of that one feature of a sentence is the only specific piece of knowledge about it that we need to know its meaning.” (p. 41)

Holism does need to be considered in a different light, because it’s unsure whether or not the view Dummett calls holism is the fairest one. The problem with an idea of meaning having to relate back to the entirety of a language is quite obvious, Dummett puts it: “there can be nothing between not knowing the language at all and knowing it completely.” (p. 44) And this is right. But is the idea of holism not that language as explained, as language of explicated meaning, that it needs more language to do so? That further justification is always carried within the language itself, and that there is no real way to atomistically butcher it up. And does it need to be “language as a whole” in the sense of langue (Saussure), why not just the full scope of possible language-use in each individual, which would then carry each individual’s ability to justify and explain what they mean, exactly insofar as their language ability reaches. I don’t see as many problems with a modified holism, since it is in line with the pragmatism of Donald Davidson, by way of Rorty, which operates under very different assumptions. This does not make it a decent or true idea by any means, but it seems that the holism Dummett talks about is a wildly naive one. It might be one held by people, but I am unsure how they would not see the obvious flaws in it. So the core notion is instead here that to determine the meaning of one sentence, you usually have to refer back to other sentences and your language-capacities as a whole, sometimes, depending on the complexity of meaning. A meaning of a word, when attempted to be explained in a social setting, requires more words, further justification in the form of added sentences. This seems far less ridiculous, to me, at least. Even with this view, of course, you have to question what it means to truly justify something, and if it’s not entirely circular and spurious to go from language game to language game. Where would the meaning finally arrive at? This is why Quine went into the observations and experience, or at least that is where I think we find the resting place of finding more meaning. The atom of language lies in the experience that prompts certain responses, you could say. After that point, we are in a very complex web of expressions and justifications, that might lead us to some kind of holism, albeit buttressed by something much more solid (experience, observation-sentences, etc.).

Dummett makes use of terms like “recognition” and “verification”, but we really do not go further with these terms than bolstering them with assent/dissent. When we recognize something as true, what is it we recognize? Where is the correspondence, what objects do we imagine we recognize? This is especially egregious when you think about the fact that each situation is unique, and the problems that come with that fact. So in reality, do we not need an explanation of what it is we verify and what it is we recognize? If this is entirely intuitive, then I am not sure we can truly claim we know what it is we are doing when we partake in verificationist theories of truth. It works, practically, obviously, without marking each of the traits that make up our recognition, but I would not call this practice knowledge of truth-conditions. Nor would I say meaning is clearly put in such a theory. Meaning comes before verification. Meaning is in my view what makes verification possible, you mean to claim something, a statement carries a meaning, this meaning itself is what is tested. Meaning can thus not be linked to the notion of truth, since truth is what applied to a meaningful statement. This subsequently makes it hard to follow anything Dummett does in his essays on meaning. But then again I am entirely clueless on the nuance of his writing, and clearly we do not share the initial starting point, since he seems Fregean at his core. Dummett nonetheless shows signs of agreeing with the lackluster nature of a theory of reference, at least: “. . . a theory of sense is needed to characterize that in which a speaker’s knowledge of the meanings of expressions of the language, as determined by the theory of reference, consists.” (p. 85) There is a requirement to explain firstly what a speaker knows in his referencing, but also how this knowledge manifests itself, that comes from Dummett’s modification of the Fregean point of view.

To answer this, with recourse to Frege, Dummett writes:

In knowing the sense of the sentence, he knows that it expresses a certain thought, i.e he knows that the sentence is true if and only if a certain condition obtains; so, in coming to accept the sentence as true, the thought which he takes as expressing represents the information he has acquired, the information, namely, that the condition for the truth of the sentence is satisfied; how that information was obtained in the first place is an altogether different matter, which belongs to epistemology and not the theory of meaning at all.

It’s curious how often Dummett uses the word “know” without thinking meaning relates to epistemology. We just assume the information is there as such, and that it can be conveyed without issue, it seems. If we are not questioning this knowledge, what can we say about a theory of meaning? Knowledge and meaning seem to be inextricably linked concepts to me, that is, if we do not assume a correspondence theory of truth or a verificationist theory of truth. And even things like “thought” and “information” are wildly vague terms. But at the same time I can’t help but think of it in Husserlian terms, where intentional objects that are synthetic and thus non-actual correspond to and find its obtainment in the actual perceptive acts. And that is where you obtain the truth-conditions that inform the noematic objects, or the virtual objects, the one we “carry” with us, the ones we use as a basis for meaningful expressions in most language-scenarios. And to be really pedantic we may have to question the idea of reference as well, in this sense. How do we ascertain the ability for one human being to convey information to another? Where a linguistic act refers back to a piece of cognitive information, how do we for certain see that it remains similar enough for recognition? To answer things like this we’d have to go deeply into the ways human beings relate to objects in our supposedly shared environment, and a notion of recognition on an intersubjective level would have to be created. How would such a thing take place? It is not enough to say that we share language, because that would be to share only the surface of what language does, to have similar enough word-sounds, etc. To say we share cultural objects and ways of living would be to get closer to it, but then you’d have to get into how we share these objects that make up the intersubjective world, or the objective world, even. To what extent are they shared? And how does it correspond with cognitive content, or information, or thoughts (to use Dummett’s terms)? If it does not correspond, what does language truly consist of?

To move on:

On either a verificationist or falsificationist theory, we should have to say that a grasp of the sense of a name consisted in a capacity to recognize whatever is to be taken as conclusively establishing, of a given object, that it is the bearer of the name. On a realistic theory, however, even this is too restricted an account: we must say, rather, that a grasp of the sense of a name consists in a knowledge of what has to be true of any given object for it to be the bearer of the name; and, since the condition to be satisfied by the object may be one our apprehension of which will transcend our capacity to recognize, in special cases, whether or not it obtains, an understanding of the name, as so conceived, will not, in general, be something that can be fully manifested by the use of the name (p. 92).

Here there is an interesting point, which has been brought up before in different words, in Dummett’s essays on meaning. “The sense of a name consists in a knowledge of what has to be true of any given object for it to be the bearer of the name”. So I know for a tree to be called a “tree”, it has to have leaves, a trunk, bark, etc., and that makes up the sense of the object which for me bear the name “tree”. But these things also need to be true, in this view, which seem besides the point, because each of these objects of sense all make up objects with their own sense, and so on ad infinitum. Perhaps that is what brings on a holism. But there might be a stop somewhere, and it would most likely stop with the inability to further see any difference between objects being described, or whatever other atomistic theory we’d end up supporting in this line of thinking.

The quote above also serve as an alternative to Wittgenstein’s idea of meaning being determined by language-use. What knowledge of what has to be true of any given object for it to be the bearer of a name consists of is very tricky. And it is not something Dummett tries to go into properly. His initial essays on meaning seem mostly to outline the shaky ground of which philosophers stand on in these matters. And ultimately, that might be worthwhile, but it does not do much more than that. Destructive theories are tedious, to some extent, even if they are important. Their relevancy, however, is unfortunately very limited. Perhaps only ones which propagate the various theories discussed here will have a proper insight into the elucidation of their respective shakiness.

In the essay “What do I Know when I Know a Language?” Dummett goes deeper into the ideas brought up in the ones dealing with meaning proper. Some distinctions between different types of knowledge are brought up, for instance one between things like learning a language and learning to swim, where knowledge of what it is to swim can exist without knowing how to swim, according to Dummett, but language cannot be known without speaking it. This is not something I agree with, since clearly you can think of what it would mean to speak a language that is foreign to you in different ways without engaging in it, the same way you can do with swimming. “To speak Swedish would be the thing you would most likely be doing if you were from and lived in Sweden,” for instance. While there are distinctions of what it means to know, there is clearly more interesting examples than this. The difference between practical and theoretical knowledge is clearly a good one, since they are of very distinct kinds, where the practical involves a more behavioristic and explicit act, while the theoretical can be kept entirely implicit but still hold firm as a mere possible behavior, or as holding the explicit aspect of it latent. This is more or less what Dummett gets at, but I am not sure if it’s precisely the same thing he is getting at. The distinction is mainly between explicit and implicit knowledge, and different ways of manifested knowledge, where one is a manifestation of an implicit already-existent knowledge, the other a manifestation of the knowledge as such, which only exists as explicit.

An interesting explanation of the goal of the philosophy of language is offered, though, one that I agree with:

What the philosophy of language has to explain is what gives this character [the significance of word-sounds] to the sounds they utter what makes their utterance expressions of thought and all these other things?

Which he then responds to with:

The natural answer is that what makes the difference is the fact that both speakers understand or know the language. Each has, so to speak, the same piece of internal (mental) equipment, which enables each to interpret the utterances of the other as an expression of thought, and to convert his own thoughts into sentences that the other can likewise understand. It thus seems as though the key to the explanation of the expressive power which makes a language a language is an individual speaker’s mastery of the language: and this mastery, as we saw, requires the notion of knowledge for its explication (p. 96-97).

The first half of this quote seems like a naive but not very uncommon view of language. What is this “internal equipment”? And how do we know it corresponds between people? Obviously this is something I ask over and over again, and it gets tiresome, but it is never answered to (as if me writing here is going to answer it, I’m sure as hell not answering it). A philosopher by the name of Georg Henrik von Wright talks about the possibility of shared “mental equipment” quite well, and came to the conclusion that there cannot be such a thing as the same pain for two different people, and this is something I’d like to extend to any sensual activity. It cannot be the same because there is no sameness to be found. There is resemblance and likeness, but that is different, and operates under different principles, and most importantly, has a lot more leeway when it comes to discussing conveying of information. But the key of expressive power does lie in explicating the mastery or knowledge we have of language, that much I can agree with. Because where else would it lie? If we want to explicate what is meaningful or expressive in an expressive or linguistic act, we have to explain what it is we know or what it is we do when we use language as a practical tool.

But as he says this about shared mental equipment, and ability to convey information and expressive power, he still wants to conflate language use with its significance or meaning. Not to say it is not for good reason. Meaning without language is expressionless and mute, it does nothing, it is of an entirely brute existence. And this is a compelling argument, because despite the possible existence of such a thing as prelinguistic concepts or whatever else, we can make no use of it, since it is outside of our sphere of practice: “thought requires a vehicle.” (p. 99) The vehicle here being language. It’s a more invasive thought, that to study thought or experience, is to study language. But this cannot be exactly what Dummett is saying, because that would mean language is self-contained, which it isn’t. Our “thought” or conscious experience seem to always point to things outside of language, objects in shared space, emotions, etc. But all the things I could list here that I claimed were outside a language, are in a language, because they have to be, because that is currently what I am doing. But if reference works as intended, if language if functional, it is functional in the sense that it can show how it is not self-sufficient, which I do not think it is. I do not think thought requires a vehicle, but I do not think “thought” is a good term. Because we don’t as such “think” a concept that is later expressed through language. The much simpler example would be experience, let’s say visual experience. It is right there in front of us before and linguistic attempt to describe or apprehend it, I think we can all to some extent assent to such a claim. And when we try to describe our experience, it is not a straight forward affair whatsoever.

Yet Dummett attaches himself to this strange notion of Fregean sense, but modified into a psychologistic sense: “The principle which Frege opposes to psychologism is that of the communicability of sense. Of some inner experience of mine, a sensation or a mental image, I can tell you what it is like. But, in the case of thought, I do not have to confine myself to telling you what it is like to have a thought that I have had: I can communicate to you that very thought. I do this by uttering a sentence which expresses that thought, whose sense is that thought, without any auxiliary contact between mind and mind by any non-linguistic medium.” (p. 102)

So there is something Dummett and others like him find very obvious in the use of language. But I can’t seem to see the obviousness. Sure there is a shared experience to some extent, in that we understand ostensive acts and references efficiently and these lead us to practical similarity or satisfaction. But satisfaction is not necessarily the same as knowledge of the same kind of the same object: “the assumption, which is, indeed, required if we are to be able to communicate by means of our utterances, that we are talking the same language, a language that we both understand: but that in which our understanding of the language consisted would lie open to view, as Frege maintained that it does, in our use of the language, in our participation in a common practice.” (ibid.) A common practice. I am not sure if I can buy the impossibility of referring to or expressing implicit knowledge. To express something implicit is not to make that implicit thing explicit and thus linguistic, let’s say to say: “I see a tree”, or “I am holding a ball”. Both of these instances of explicit language are not at all the same thing as actually doing those things. We can even express the difference linguistically, but only by reference to something not inherently linguistic. We have to understand this difference between sense-experience and linguistic acts, it is very clear. And I do believe we may have to take the issues that come with such a decision to allow for private concepts. It would be absurd to reduce pain to explicit or expressed pain, and it would be to forgo a large aspect of experience to only let that which is intralinguistic be discussed. But perhaps there is an extralinguistic aspect to this common practice. Perhaps the commonality of objects is also a thing, and the commonality of pain, due to shared “mental” equipment, but we have no way of determining such a thing, as Dummett also writes. So what is it that we share, what is common to us in language use? The form of the language? The prompting of similar responses in similar circumstances? Ultimately Dummett seems to claim, in the essay on “What do I Know when I Know a Language?” that there are answers to these questions. And that language use, and its subsequent aspects that are knowable, are available to reflection. A lot of his concluding words are words I agree with:

If this is right, it follows that the notion of knowledge cannot, after all, be extruded from the philosophy of language. It has also a further consequence for the criterion of success in constructing a theory of meaning for a language. For it follows that such a theory is not open to assessment in the same way as an ordinary empirical theory; it is not judged correct merely on the ground that it tallies satisfactorily with observed linguistic behaviour. Rather, the only conclusive criterion for its correctness is that the speakers of the language are, upon reflection, prepared to acknowledge it as correct, that is, as embodying those principles by which they are in fact guided. Such a theory cannot be arrived at by observation alone, but requires reflection; and it is by reflection that is must be decided whether it succeeds or fails (p. 105).

Again, the problem is shifted, and we are left somewhat unsatisfactorily with the notion of reflection as a “what?” I am not sure who it would be clear to, through reflection, what a theory of meaning could be. In reflection alone, in its most radical form, which would be a form of phenomenology, we do not find answers that lead to a genuine practical language. But that more often than not, if we can ascribe people like Deleuze with a type of radical empiricism or phenomenology (they are different, but can sometimes have similar results in their applications, it seems), leads to language not having its intended function. Reference does not work as intended, since we can always say we see the two aspects of reference, let’s say word and visual perception, but never the link between them. Husserl claimed we could, but it is not at all a vivid experience of the same kind, and clearly not self-evident, which would to some extent leave it open to the later Husserl’s epoché.

In the essay “Language and Truth” Dummett tries to explicate the relationship between language-use and its deep connection to truth in a more rigid sense than what has been attempted in the earlier essays. Firstly he takes up the idea of “thoughts” again, in their full vagueness (thoughts about what?). Sentences are expressions of thoughts, or better, thoughts are what different language-acts stand for. “We could not grasp what it is for a sentence to be true if we did not regard it as expressing a thought and understand what it is for a thought to be true.” (p. 128) The truth of a thought seems simply determined by recognizing it as such, a thought is recognized as true or false. This recognition, however, does not seem to be explained in any real manner. This is all with the help of Frege, and it is sometimes unclear where Frege ends and where Dummett begins. All we get in ways of explaining recognizing something as true is that it consists solely in assertoric statements, that is, in explicit sentences.

There is an interesting distinction created in all this confusion, though. Truth-predicates as they relate to what language stands for and the sentences themselves:

True” in the sense which Frege considered primary, as a predicate of what sentences stand for, is expressible in his formal language, and, even if not taken as primitive, would be definable in it. “True in what he considered the derivative sense, as a predicate of sentences, is not expressible in it, principally because the language does not contain a means of expressing the relation between symbols and what they stand for.

This to some extent echoes what I mentioned earlier on the notion that it is not possible to within language validate language. If language is not about something, or does not stand for something, it does nothing whatsoever, and no truth can be gathered from its sentences.

But to avoid the circularity of a holism, and to still find an extralinguistic aspect of linguistic acts, Dummett comes to say:

[A]n explanation of the meaning of a sentence may presuppose the meanings only of sentences of a lower complexity, and will perhaps be given simultaneously with the explanation of certain sentences of equal complexity: it will never involve explaining or presupposing the meaning of any sentence of higher complexity. An understanding of any sentence will involve, on this hypothesis, an understanding of some fragment of the language, a fragment which could, moreover, exist as a language on its own; but an explanation of the language as a whole could be constructed without circularity by starting with sentences of minimal complexity (the observation sentences) and completing the explanation of the sentences of any degree of complexity before proceeding to the explanation of those of the next degree (p. 139).

This mirrors (excuse the imagery) wonderfully the acquisition of language, and this is very important. And I do not at all think this would be something Quine would object to awfully, which is something Dummett thought. A holism based in originally atomistic sentences, that started in observation sentences, thus in sense-experience, is something very valid to me. In this sense, we can only find meaning in one sentence by referring to another, and ultimately, by referring to objects in shared space. What complexity here means, though, is unclear. If degree of complexity is the same as degree of elaborateness, then I might have some clue on what is going on here. But we still have to refer to extralinguistic behavior, such as pointing, and the implicit grasp of such acts by children. So the atoms of a holistic language would be just these things, the ostensive acts, the observation sentences and what is built upon this foundation. Obviously the problem still exists on what it would mean for an observation sentence to correspond with the respective sense-experience, and how they relate to each other when the behavior is not there. What does the link consists of? To invoke Husserlian intentional objects of reference and association would be an easy way to go about it, but it would leave us open to the vagueness and difficulty of actually obtaining such an experience of linkage. The experience in a natural or real sense would be the one of pointing and being directed towards salient points in space, in the way described by Quine. If we suspend skepticism for a moment, it is hard to see the flaws of such a system, since it seems to run in tandem with how language is acquired and how it is used in everyday life. When you are confused about the meaning of something, what resolves the confusion usually ends up being a demonstration of some kind, which in other words would mean forming an observation sentence through an act of observation. “This. . . is what ‘Red’ is”, “This . . . “ being the ostensive act and all the implicit aspects involved in such an experience.

To keep constructing such a notion, we could form almost a natural Husserlianism. The intentional objects as such would simply be the observation sentences’ objects, and their directionality, this directionality then being represented by the ostensive act. This directionality is clearly carried with us beyond the ostensive act itself, which make the object of our directionality outside the actual act something internal to us. This would, it seems to me, mean that there is some kind of intentionality in the Husserlian sense, which should not be denied in general, but I know it is by many philosophers of the Wittgensteinian tradition (the later Wittgenstein, that is).

But in this view of lower complexity sentences to higher complexity sentences, Dummett gets closer to an idea of meaning and verification. A more complex sentence finds its meaning in lower complexity sentence. And you verify the meaning of a sentence by looking into what a full sentence’s constituents are. “A direct verification of a sentence like “There are nineteen eggs in the basket” will consist of counting.” (p. 142) So here we have a structure based on observation sentences on a very basic level. And verification comes about through understanding what goes into creating a complex sentence, and especially what goes into its lower level sentences. I may be extending Dummett’s view here, but you obtain the meaning of things like this through a phenomenological analysis, really. That is exactly what is going on here, in my view.

Ultimately this view might just be possible to work with, but it may not be unproblematic enough to fully develop. And Dummett, in his version of it, is still not ready to adhere to any of the possible theories he describes. Yet again focus is shifted away from solid answers, and the focus is on destructive philosophy and lamenting the inability to come to a conclusion:

I am not asserting that the notion of truth, as employed in the theory of meaning, cannot be wholly explained in terms of those of verification and of consequences. I hope very much that it will prove to be able to be so explained; that would constitute a final resolution of the philosophical problem of truth. If, however, the notion of truth can be explained in terms of those notions, it can also be replaced by them, and would be better so replaced. More exactly, in such a case a truth-conditional theory of meaning could, and preferably should, be replaced by one in terms of verification or of consequences. If this cannot be done, we have the alternative of accepting, as essential to an account of our understanding of our language, a notion of truth which in principle resists complete elucidation, or of admitting radical error in accepted modes of reasoning. Perhaps fortunately for our peace of mind, we are not yet in a position to say whether we need face this choice or not (p. 146).

In “Truth and Meaning” Dummett continues his attempts to respond to what meaning, truth and language-use is all about. He gets closer to a behaviorist view this time, similar to Quine in fact, in that he refers to the situation where a child is taught to use a word or phrase in the “right” circumstance. But what is unsatisfying in such a reference is that we still do not have any sense of what is the connection between what is said and the supposedly correct circumstance other than, again, the word “recognition” and assent from a parent. Surely, again, and again and again, we need to go into what recognition is and how language at all can evoke these different responses, and how language is even connected to the state of affairs at hand. When we talk about it in this roundabout way we do not actually explain what is going on with the people involved in these games.

But it seems as if with each essay, Dummett gets closer to a proper statement regarding the issues and guidelines of our problem: “A theory of thought must give an account of whatever is involved in the activity of thinking; a theory of meaning must describe the practice of speaking a language.” (p. 159) But the activity of thinking, or mental acts, on their end is explained through their mental content, which are the thoughts, and the sense of sentences. So in what way is this not just a circuitous circularity? What do we actually gain from these movements back and forth between terms? I don’t see it. The wonderful thing about it, however, is that it apes after the interesting aspect of Davidson’s philosophy, again, as I wrote about above, in that it wants to make the description of practice a vital part of meaning. And this is more than a mere “meaning is use” kind of ordeal, it is a self-reflexive source of knowledge, a direction towards one’s own actions and usage as explanation. But the question turns around on this entire situation as well, and we have to ask how the language we use to explain these things work, still, for the explanation of our practice – which comes about linguistically – can hold any power. And to go further, it also applies to what I am doing now, in writing. Isn’t it tiresome?

According to Dummett, it seems as if he does not want to be interpreted in a “crudely behavioural manner.” (p. 161) At least he does not want the specific theories he is trying to ground to be seen as such. “That is to say, you could not, by observing the speakers of a language you did not know, decide what they treated as justifying the assertion of any given statement, or what they regarded as involved in accepting it: you would have to know quite a lot of the language before you could discover that.” (ibid.) So if this is the case, what is the point in appealing to language acquisition scenarios and things that have been talked about above? An argument against this would be the easy question of: How does a child acquire language if it is impossible to penetrate from the outside? Are we going to pretend that thought and language are so similar that there is a straight forward move from the one to the other? If thought is highly linked up with language, now using Dummett’s terminology for the sake of it, how would we ever acquire language? If language cannot be explained through the non-verbal signs of intersubjectivity, how does it arise? What internal process leads to its manifestation? I don’t see how language can come from anywhere but the outside. Surely we have a predilection towards acquiring it, since we can and other beings cannot. This seems more like an ability to understand non-verbal communication intuitively to a higher degree than other species, than it seems to be a sign of some kind of innate language springing out of nothing. Clearly this is not the case. So what is Dummett getting at?

Sure, justification is a complex process which require a deep knowledge of language. But lower level justification, which comes about through simple reference, is surely a behavioristic process. The justification of calling something you perceive as red as “red” is not justification in any philosophical sense, but simply an establishment of rules, guided by parental decree and assent/dissent. Perhaps it could be claimed that this lower level language is not justified in its use. But if this is the core of language, how is what is built upon this acquired language justified whatsoever? I am clearly confused in some kind of way.

In “Language and Communication” the notion of thought becomes more prominent. Finally Dummett provides the reader with some notion of what thought is, and how it can be properly seen: “For a thought does not resemble a mental image or a sensation: it has the distinctive feature of being, or at least of being capable of being, true or false, and thus relating to reality external to the mind.” (p.170) So thought is thus a relation to an external event or object, that is, external to the mind. This relation seems to come either from a direct relation or an indirect relation by way of an object in external reality causing a mental image which then represents the object that caused it. Dummett does not think this explanation suffices for them to carry a truth-value, however, which should be clear to anyone invested in these philosophical problems. Obviously objects change around the mental image we have of them, if we can now assume that mental images are one-to-one correspondences with objects, which is uncertain. And then we can even question the notion of imagery being possible mentally, as if the mind carried photographs, which is clearly not the case either if you just spend some time with introspection. The word “image” is wholly unfit for a discussion like this, since we are not dealing with mental images. This doesn’t even take into account the antirepresentationalist views of people like Rorty and Deleuze which would entirely break down a correspondence theory of truth before it could be established.

Dummett moves on from a representationalist view as such, and talks of language as expressive, as opposed to coding thought. This is an attempt to make language closer to what it refers to and its meaning.

There is a fundamental difference between expressing a thought and using some conventional means to identify it. Given an invalid argument, the phrase “the weakest additional assumption needed to render the argument valid” picks out a unique thought; but it does not express that thought, since it is possible to understand the phrase without knowing which thought it picks out. A sentence expressing the thought, on the other hand, cannot be understood without knowing what thought it expresses. It is an essential feature of anything properly called a ‘language’ that its phrases and sentences genuinely express their meanings. That is the difference between a language and a code; and that is why the mastery of a language enables a speaker to grasp new thoughts expressed in it.” (p. 173)

The meaning for Dummett of language lies in knowing what thoughts are being expressed, then. Meaning seems to be the respective thoughts that are expressed in any sentence and its content. In following Frege, however, sense is always expressible fully. A difference in sense also means a difference in language, as when two people use the same proper name but have disparate sets of knowledge about the object or person in question. Let’s say we are looking at the same square, but on your side it is yellow, in my side it is red, we have a conception of a cube in front of us, and can call it “The cube in front of us”, but what constitutes its sense differs to the point where we are not truly speaking the same language, in that language is constituted always by its content, thus its set of senses.

In the essay “Frege and Husserl on Reference” Dummett discusses the difference between Frege’s and Husserl’s ideas of reference. For some reason Dummett is motivated to conflate the two. Specifically the Husserl of the Logical Investigations is what Dummett thinks can be seen as similar to Frege, when it comes to reference. But the difference is simple to state. Even in the Logical Investigations, Husserl avoids hierarchical existence, and it does not seem he even wants to admit such a thing as non-existence. “Everything that exists can be experienced” is the phrase that resonates most strongly in accordance with this. Frege and Husserl have similarities in that sense and intentional objects are apprehended somewhat similarly, they are at their core almost descriptive. And they both refer to specific objects. But when we think of the term “object” between the two philosophers, we have to really look at the nuance. It was a while ago since I read Frege, but if I recall correctly, his reference is in regards to actual objects in the world. Whereas this would be to say too much for Husserl. The objects of Husserl are not objects in this naive sense that Dummett conflates his objects to be. There is an intended object, and this is what determines an experience’s materiality (again, try not to think of this in the naive ontological sense), but all of these objects are of equal validity in experience. So even if you think of something that cannot be actually or directly perceived visually, you cannot claim that it does not exist, but what you can claim, according to Husserl (and I agree with this), is that there is a fundamental difference in quality of experience. Direct experience is self-evident, and it comes closest to the intended object, since it’s the first-hand experience of the object, but this is by no means the only way of apprehending or experiencing an object, or even the same object. I think the fundamental difference, though, between Frege and Husserl’s references, is the depth of which they go into it. Husserl is not content with simply thinking that the object is what is referenced as such, but explains what the object is in great detail and how it is synthesized through experience. Frege is more straight forward, and does not avoid ascribing ontological attributes to experience. Husserl’s notion of experience is all there is to the world within Husserl’s experience, since the fundamental question is what knowledge is and how it finds it foundation in experience and is then developed and manipulated from that point on. Reference then, is also something that exists as an intentional object, which I don’t think is something Frege’s philosophy could arrive at. Then again, I do not know Frege very well, so all of this might not be particularly worthwhile. In any case, Husserl’s reference is not exactly a reference to the world as such, but a reference to meaningful intentional objects that are apprehended adverbially, or through description. It’s a description of experience we are dealing with, and thus a reference to various acts and objects that are presented to us; but nowhere in experience do we find non-existence or a sense without reference, because in Husserl, what the equivalent of sense would be, could be referred to as an object of experience. Hopefully this would be some kind of elucidation of the difference between the notion of reference in the Husserl of Logical Investigations and Frege’s conception of reference to objects in a very ordinary sense, unmediated or unproblematically.

At this point I’ve lost all interest in writing as a starter, and likewise all interest in Michael Dummett, and thus especially writing about his philosophy. But in Testimony and Memory, I found the reason why. He is trying to escape cognitive solipsism. Dummett wants to uphold the past and future as proper aspects of knowledge. But he does not seem to want to discuss what is genuine knowledge and what isn’t, all knowledge is just knowledge. It’s tiresome to write about for that reason, since there is no nuance in a term like this. Either we know everything we claim to know, or we know only too little for knowledge to be a term worth talking about. But is this not a fine position to end up in? We would have simply just came to a point where the incessant philosophizing lead to something. To me, if knowledge can decay, it is not genuine self-evident or “proper” knowledge. Our relation to the past is very distinctly a special kind of cognition, and I think Dummett knows this, but wants to uphold its relevance, for some reason. He is under the impression that if we cannot hold the past as proper knoweldge, we cannot have a self, that we cannot almost know who or what we are. He does not want to reduce us to momentary subjects. But many would claim that is exactly what we are. Our most genuine and true relation with a world is through the direct and actual experience of it. That is where we look when our other forms of knowledge falter. Clearly things are changing, and the past is different from the know, so what kind of knowledge is accurate, that is based on the past? We find it in constants, we find it in the unchanging. And all past experienced are unchanging and true in and of themselves as exactly what they are. But it seems that Dummett wants to carry the past into the moment, and want there to be a correspondence between the two, which there simply is not in the vast majority of cases, despite the fact that we do act as if that is the case. It is most practical to approach things this way, and it is an approach that has been bestowed upon us simply through our capacity to apprehend things with a memory.

But if we are to follow Husserl’s kind of knowledge, what makes these objects of knowledge possible, and genuine, is the ability for us to obtain them again and again through actual experience. So you say, “there is a tree in my backyard”, this statement once said when you are not there, is uncertain, we move away from our backyard, and talk about our tree in other scenarios. If we come across a skeptic, that asks us to prove that we have this tree, we take this skeptic to our backyard and show our tree, which then leads to dispelling of any doubts, hopefully. We do this exactly because a memory is never of the same quality as an actual experience is, and because we can always direct a fair amount of skepticism towards memories, whereas the same is not true for direct experience (if we don’t go overboard with the Cartesian skepticism, which I do not). At the same time, knowledge seen as apprehended cannot be a momentary thing solely, since that would make it impossible to say you knew anything in a real way. But knowledge is not contained solely in the memory alone, it is in the possibility of obtaining its meaning again, through experience, that we find knowledge. Our language that directs these is part of what makes up knowledge, our mind, our sense of directionality, and so on. A characteristic of knowledge is that it is obtained and apprehended, thus carried with the person. And that might be the fundamental problem with knowledge that claims to be about something not within its sphere of apprehension. To be fair, I am not myself sure yet on how to approach knowledge, but I am not so quick to defend a position I have no way of defending other than with an appeal to “oughts”.

This also leads to distinguishing between two kinds of knowledge of past events (there are probably more):

In one sense there is the unique events of the world. These cannot be apprehended again, and their knowledge seem to consist in mostly the linguistic description of them. Knowledge of this kind of non-demonstrative and thus, I think, should be of a lower kind, a less certain knowledge; although nonetheless it is important and very useful.

On the other than there are predictable events, which would then mean that events are repeated to some sense. The repetition in question is constituted by similarity between events. In autumn, leaves fall off the tree, for instance. We identify the similarity between individual trees, we identify the season of autumn as that which comes between summer and winter, and whatever signs make up the signs of autumn, and we thus form a prediction of past knowledge, a matter of fact, in the Humean sense, about deciduousness. Each repetition is an affirmation of the description, and it is demonstrative.

In short, one is possible to corroborate, the other is not. Or, better yet: One is possible to corroborate only through sociality, through communication (the former), the other you can obtain through observation alone. But the question is if this distinction is genuine or not. Since in some sense I’d like to follow people like Hume and Bergson in saying that there no way of apprehending the future through the past, and also that each event is in some way unique, if only temporally speaking. This, I think, puts a particular question forward on what it is knowledge of this kind actually does, or what it is directed towards. And I suppose both Husserl and Quine have their answers, to some extent, that they are directed towards either objects, material objects in the Husserlian sense, or salient qualities, in the Quinean sense.

I have a great deal of philosophical concerns that Dummett simply does not respond to. He does a lot of great analyses, in almost a Merleau-Pontean manner, where he does not strive too far into either “camp” that is being discussed. He tears down, but I am unsure how much he ever constructs. This leaves a lot to be desired as somebody who is looking to learn. Of course, a lot went over my head, I’m a worthless idiot, but in the essays that did engage me, I found nothing to properly hold on to. Dummett will not be a person I’ll read more of, most likely. I shall direct my efforts to things that affect me more greatly, and have a more constructive quality to them. Or maybe I’ll just do whatever comes to mind, maybe I’ll torture myself more in the future. Since this is, in fact, all an exercise in intellectual and emotional torture. A perpetual sense of stupidity and inability to hold on to any sense of security. Philosophy is a doldrum when you read as I do, and jump from thinker to thinker, who all repeat the process of coming to a conclusion, most of which end up in a wildly dissatisfying place. And I know I am doing the exact same thing going through these thinkers, subjecting myself to thoughts that bear no impact whatsoever on anything whatsoever, barely even on myself and my intellectual process. But on we go.

(Also, sorry for the repetition and dull writing, I hope nobody reads this.)


One thought on “Dummett’s The Seas of Language and my Moronic Skepticism.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s