Alice: "I don't know what you mean by 'glory.'"
Humpty Dumpty: "Of course you don't -- till I tell you. I meant 'there's a nice knock-down argument for you!'"
Alice: "But 'glory' doesn't mean 'a nice knock-down argument.'"
Humpty Dumpty: "When I use a word, it means just what I choose it to mean -- neither more nor less."
Alice: "The question is whether you can make words mean so many different things."
-- Lewis Carol, Alice in Wonderland
Introduction: What is Ordinary Language Philosophy
I take Ordinary Language Philosophy () to be any philosophical program that claims that the problems of philosophy are best solved (or, in some cases, completely solved) by paying attention to ordinary language. As the Wittgenstein of Philosophical Investigations says:
Philosophical problems arise when language goes on holiday.
(Though Wittgenstein spawned an era of , it is arguable whether he himself was a through-and-through philosopher.)
An example of tackling a philosophical problem using might be as follows. We might wonder "What is triangularity?" Sure we see triangles, but we don't "See triangularity". In comes the philosopher to insist that looking for a "thing that we see" in place of triangularity is a mistake. We should, they will say, pay attention to how we use the word in ordinary language, and we will see that we have made what Gilbert Ryle would call a category error in assuming that our singular term "triangularity" should refer to something like "That balloon" does. They will say that we can avoid being so misled by sticking to our ordinary language guns and not letting language "Go on holiday".
In this short note, I hope to outline where I think goes wrong and why, and also what it gets right. I will argue that is correct in the following points:
You can't get semantics before pragmatics. (Don't look to the "meaning", look to the use)
Theoretical meanings have to be related somehow to pragmatic meanings or they are useless.
If meaning comes from use, then meanings will be an imprecise hodge-podge because "doing" and therefore "use" is an imprecise hodge-podge. This is why "conceptual analysis" has never successfully analyzed a concept in the history of philosophy.
If meanings come from uses, then they must follow rules. For communication to be possible, it must be normative.
and I will argue that is incorrect and should be polished up in the following points:
Ordinary use has no theoretical primacy. A term can grow to encompass more than its ordinary use.
What claims ordinary language sentences should/shouldn't, do/don't commit us to are not known implicitly. They must be accounted for theoretically and explicitly.
Not all philosophical problems are dissolved in ordinary language as it evolves and keeps up with science.
I think is an immensely powerful tool and like all powerful tools, it is prone to overuse and misapplication. Although I once believed it, I have long since abandoned the idea that all philosophical problems can be resolved by appealing to ordinary language.
What Does Ordinary Language Philosophy Get Right?
You can't get semantics before pragmatics. Don't look to the "Meaning", look to the use.
We cannot pick the meanings of our words out of thin air. So much the worse for Carol's Humpty Dumpty. They depend on learning sets of concepts. It often seems like philosophers are doing this when particular philosophical puzzles arise. An example that tends to bother me is when philosophers fall prey to the notion that we can sort of "mean things at will". They do so when they say, for example, "Maybe the red I see is what you see when I see blue!" How exactly do we mean "red" in that sentence? In the same way that we learn how to use the word "red"! And, if that is the case, then it seems like I cannot be meaning it "in some other way at the same time" that presumably only I could understand. Or if I am, then that way of meaning that just sits flaccidly alongside the pragmatically efficacious one (the one that has cash-value of meaning in our actions) and isn't really a part of communicating with others.
It is as if when I uttered the word I cast a sidelong glance at the private sensation, as it were in order to say to myself: I know all right what I mean by it.
-- L. Wittgenstein, Philosophical Investigations 274
I don't have preconceptual, in this case prelinguistic, knowledge of what "red" is and so how could I possibly mean something by it that is directly out of accord with the pragmatics of that concept (How I learned to apply it). I might as well say, "maybe the feeling I get when I take a bath is the same as the one you get when you have a toothache".
is correct to steer us away from these illusory meanings and the knotty philosophical problems that come with them. I think it is correct, not by virtue of the primacy of ordinary language, but by focusing on what it is that gives our expressions meaning in the first place.
Theoretical meanings have to be related somehow to pragmatic meanings or they are useless.
When we are giving some kind of theory, predictive, descriptive, synoptic, or prescriptive, the theory will have no use to anyone if it is not tied to a concept that has pragmatic value to us. I think the free will issue is illustrative of this kind of divorce of theoretical meaning from pragmatic meaning. My typical argument with the determinist who acts like they are the first person to discover that the physical world of science is causally closed goes something like:
Me: "So if you signed a contract today, you wouldn't be responsible for that contract?"
Determinist: "No because I never chose that. The particles [etc] just caused my hand to write like this and that and my upbringing caused me to be thus and so [etc etc etc]"
Me: "So no one REALLY ever makes a choice? Really we are just misapplying the concept of 'choice'?"
Determinist: "Yes! There are no REAL choices."
It should seem peculiar to us that "real choices", on the determinist's view, turn out to be just the sort we don't have. The pragmatics of what is and isn't a "choice" have gone entirely out the window and with them the notion of what we are talking about having any material consequence. And so the conversation will circle back to "What is a choice?" and the determinist and compatibilist will disagree on that. But why should we care about this particular determinist's definition of choice? It seems like they have a big "So what?" to answer for next if we accept it. When we divorce theoretical "choice" from what practices and actions we take in knowing and deploying the word "choice" in real life, what importance does the theory really have?
is correct to keep us on a slightly tighter leash than we sometimes want to be on when philosophizing. We need to stay somewhat tethered to the common-sense concepts if we want our philosophizing to pertain to them.
If meaning comes from use, then meanings will be an imprecise hodge-podge because "doing" and therefore "use" is an imprecise hodge-podge. This is why "conceptual analysis" has never successfully analyzed a concept in the history of philosophy.
We should assign an extremely low probability to the possibility of 'fully analyzing a concept'. In an effort to make the concept precise and immune to counterexamples, we will confine the concept to a tiny playpen and thereby commit the mistake of divorcing the concept from the vast landscape of actions and practices that give it meaning at all. We either have a chaotic hodge-podge of a concept, a fragile one to counter-examples, or one so surgically defined that it simply has nothing to do with the concept we sought to elucidate in the first place. I believe accepting the hodge-podge is the best play here.
That doesn't mean we can never have precision in more tightly bounded domains. It only means that precision across the board in every practice and context, as the determinist above thinks that can have with "choice", will likely be impossible and should not be expected. If meaning is determined by what we do, and what we do changes, how can we expect to ever catch up?
If meanings come from uses, then they must follow rules. For communication to be possible, some normativity must be assumed.
As Wittgenstein (under some interpretations) and Kripke point out, in their rule-following paradoxes, applying a concept properly, learning a word, following a rule cannot be something that is done all by oneself in a vacuum. Kripke asks us to imagine a rule for a mathematical function called quus. Quus, , is just like 'plus', , except:
if
but
if
Suppose I have been teaching you to add all the way up until we have gotten to:
What should you answer? ? or ? The point of the strange example is that it seems like nothing about how I have taught you to add so far has determined whether I meant quus or plus. There is no fact of the matter. The same skeptical challenge could be applied to any rule and so it seems that without some turning of the spade and reaching justification for the rule, this could go on ad infinitum. We reach a sort of Agrippan trilemma of "rule following". It seems no rule can ever be justified merely by example or stipulation of how the rule should be followed because any action can be made to fit the rule and any rule made to fit my past actions. If we need rules for how to follow the rules, then how could a rule ever ground a practice? Wouldn't we then need "rules for following rules for following rules"?
I think the rule-following paradox terminates in doing. We don't learn rules by painstakingly memorizing instructions or "rules for following rules". It isn't that doing is a kind of justification. It is that we turn implicitly on certain hinges when we act. Our actions implicitly commit us to taking certain explicit, though not practically codifiable, claims to be the case. This is were the regress ends.
Someone says to me: "Show the children a game." I teach them gambling with dice, and the other says "I didn't mean that sort of game." Must the exclusion of the game with dice have come before his mind when he gave me the order?
-- L. Wittgenstein, Philosophical Investigations 70
In this silly example, when the concerned parent says "Not that kind of game!" he didn't consult an inner rule or mental lookup-table of any kind in order to exclude gambling from the kind of games he meant. The meaning of the sentence and the rule that we were to follow was grounded in practice, in the pragmatic "doing of stuff", in habits and the innate nature of us as human beings.
Ordinary language is rough around the edges, does a million more things than state facts about the world, constantly changes, and is a reflection of what we do and what we fundamentally believe must be done. There is no playing the game of philosophy without language or concepts of some kind which set their roots in practice. Therefore, we have to philosophize while keeping in mind that those words and concepts are constructions of the practices and actions that give them meaning in the first place.
What Does Ordinary Language Philosophy Get Wrong?
Ordinary use has no theoretical primacy. A term can grow to encompass more than its ordinary use.
I am often frustrated reading philosophers say things like "This is a misuse of the word thought". I am all for taking the semantics as following the pragmatics. We cannot just mean whatever we want by words without getting the ball rolling through some kind of action in practice. If I want "thought" to mean "words I speak internally", then I am certainly allowed to adopt that as a technical term. It may not relate very well to the ordinary use of the word or concept, but maybe that isn't my goal in whatever particular mode of philosophizing I am doing.
philosophers have attacked neuroscientists' use of sentences like "The brain thinks X". They claim this is a misuse of "to think" as only persons can think. That's where the pragmatics come from they claim, and I agree. We learn to say "think" of someone based on how they are acting etc etc. However, I think it is silly to claim that this kind of metaphorical talk is "nonsense" full stop. We very frequently use intentional language as a proxy to quickly understand complex systems and that should stand on its own theoretical merit, not be judged by the stipulated authority of ordinary language.
What claims ordinary language sentences should/shouldn't,do/don't commit us to are not known implicitly. They must be accounted for theoretically.
There seems to be a sentiment in that all philosophy should do is describe how terms are used. However, in reading some of these philosophers, Ryle particularly, I often noticed that they seemed to be doing a lot more than that. More specifically, I often notice that they argue for or take for granted certain theories of how we should account for a given way of speaking.
These supposed "mere descriptions" from philosophers are often just expedient theories of how something in the world works. Sometimes they are empirical and need evidence, sometimes they are empirical and don't need evidence (because they are so obvious), and sometimes they are synoptic/prescriptive accounts of how language should be viewed to work. Here's an example: Quine (another philosopher who puts pragmatics before semantics) gives the example of "sakes" as a word that takes the grammatical form of a singular term but does not refer to anything. There are no "sakes" to be accounted for. Did we automatically know that there weren't "sakes" by virtue of knowing how to say "for Pete's sake"? Is that question "nonsense"? Is Quine giving a theory of how our words should be viewed? I think he is and so is every philosopher who prompts us to focus on the ordinary use of a word to circumnavigate a philosophical problem. The theory is still a theory whether we got it from observations about ordinary language or from philosophical accounts after the fact. We cannot know all there is to know about the application of a concept merely by looking at its current use because that concept's application is a prescriptive, not a purely descriptive endeavor. Should we say "it is the same time here as on the moon" in a post-relativistic physics world? Ordinary use cannot give us the answer. In a sense, our concepts are in a constantly changing state. We are almost always engaging in some small amount of conceptual engineering every time we encounter a novel case. But don't get me wrong. I'm all in favor of eschewing "accounts of" something when the ordinary concepts will do just fine.
We don't know all the implications of our claims just by virtue of making or being disposed to make them. Additionally, we don't always know how the ordinary concepts we use hang together, or how they relate to one another. We can be masters of the concepts of "inference" and "belief" without ever having thought about how they relate.
The pragmatic move is a good move, but it does not instantaneously give us a correct and satisfactory account of how our terms work, how their use can be more technically refined, or how they do/should relate to the word as science tells us it exists.
Not all philosophical problems are dissolved in ordinary language as it evolves and keeps up with science.
Even in with a strong adherence to not extending ordinary terms beyond their common use, it isn't clear to me why all philosophizing is off limits by virtue of it not having some property that ordinary language has. I'd agree with Wittgenstein that lots of philosophical problems arise from our "bewitchment by language" (non-referring singular terms being a good example), but it is far from clear to me that we can say all of them are.
Philosopher David Enoch has a very fun example to illustrate this. In Pulp Fiction, Vincent Vega and Jules Winnfield are having a little back-and-forth about why Jules doesn't eat pork.
So by that rationale, if a pig had a better personality, he would cease to be a filthy animal. Is that true?
Enoch points out that both speakers are clearly philosophizing. It seems wrong to me to say that they just crossed the precipice of sense and were talking nonsense to each other even though they thought they weren't. On the other hand, I do think it is possible to philosophize in vain by playing games with words that are too far gone from their practical application for the inferences we might validly make about them to have any relevance at all. I take philosophizing to be a general extension of the basic human discursive practice of making our commitments to certain claims clear, whether to ourselves or to one another. That's what Jules is doing, and that is what I am doing now.