The Analytic:Synthetic Distinction

[Music] hi i want to talk to you today about analyticity in other words the analytic synthetic distinction it's one of the basic distinctions of philosophy it's primarily a semantic distinction and i want to look at a little bit of its history but focus on how it's understood today so first of all there is something like this in aquinas when he talks about things that are true because the predicate is contained in the subject what he really means is that there is one concept that is contained in another concept leibniz later developed an entire calculus of concepts it talks about containment as the basic relationship among concepts this notion gets developed and really recognized independently for the first time by john locke who refers to certain propositions as trifling propositions propositions that tell us nothing about the world but may express something about logic or about the nature of language so for example if i say a soul is a soul that isn't a very helpful answer to the question what is a soul but it is true a soul is a soul lead is lead water is water and so forth those things are logical truths and some of these are truths by definitions he thought that lead is a metal is a good example of this or that gold is yellow those aren't great examples in my view but there are other kinds of examples things that would be true by definition like bachelors or unmarried for example if we reflect on logical truths and truths by definition we can ask what they have in common and the answer seems to be this their truth value is determined by the meanings of the terms inside them a soul is a soul or bachelors are unmarried true if you just understand the basic terms like is or bachelor that's all you need everything else follows from language so we might describe these as purely verbal truths truths that don't tell us anything about the world and i'll come back to the difference between those two ways of understanding analyticity in a moment the terms analytic and synthetic were first introduced by immanuel kant kant says that an analytic judgment is one where are getting back to aquinas the predicate is contained in the subject the concept of the predicate that is is contained in the concept of the subject we can understand that that way but almost immediately people generalize that to thinking in broader terms it's a matter of having your truth value determined by the meanings of the terms so we can say that a judgment is analytic if and only if the meanings of the terms within it determine whether it's true or false a synthetic proposition is simply one that is not analytic so it's synthetic if the meanings of the terms do not determine its truth value to do that we would have to presumably look at the world the classic attack on the analytic synthetic distinction was given in an essay two dogmas of empiricism by willard van orman kwine quine's argument is complicated and i think there are many different interpretations of exactly what he's doing in that essay here i'll just say that he thinks that an empiricist really has no right to that distinction now why not well essentially he says to understand analyticity you have to understand meaning it's a matter of having your truth value determined by the meanings of the terms but how can you understand meaning he thinks that meaning is not something that we can really have access to in experience and so he doesn't see how an empiricist can really draw this distinction given that they really don't have a grasp of meaning now he said that problem would be solved if the empiricist had a grasp of necessity then we could say there's a kind of verbal necessity to these truths like a soul as a soul or bachelor was run unmarried but he ends up saying there is no way to grasp necessity hume taught the empiricists that he thinks and so in the end he thinks that analyticity meaning necessity synonymy that is to say things having the same meaning and a variety of other concepts are all tied up together and they're all problematic for an empiricist that's not going to stop the rationalist from drawing the distinction but it will stop he thinks an empiricist and so he argues for a version of empiricism that does not rest on the analytic synthetic distinction doesn't take some of the truths the analytic ones and put them aside and say i'm really talking about the others as depending on experience for quine everything has to depend on experience another way in which he seems to object is that there are certain propositions that it's very hard to classify his example in the essay is everything green is extended that is to say extended in space and time has some spatial extension some width some breadth um is that true by definitions of the terms he says i'm not sure there was a similar debate going on at the time among various philosophers who are contemplating propositions like this nothing could be both red and green all over not i'm not talking about a red and green plaid or something like that i mean red all over and green all over well that seems true nothing could be both red all over and green all over but what kind of truth is that is that something we learn from experience is that something that's true by virtue of the meanings of the terms when we reflect on it we realize we can't intuit it we can't grasp or imagine or even conceive of maybe a situation where something is both red all over and green all over but are we willing to say gosh that's true by virtue of the meanings of the terms we'd have to understand the meanings of terms like red and green and it's not clear that we understand them in a way that would make them incompatible another objection that quine seems to raise in the essay is how do we end up drawing the distinction we can sometimes give a semantic theory that is to say a theory of meaning but that is itself an empirical theory that is to say it rests on experience well if that's true then you might say the very basis for declaring certain things to be truths of semantics truths by virtue of meaning to take certain things as semantical rules is itself part of a theory that rests on experience that would be a way of focusing on what i call a boundary question not saying something like is everything green is extended analytic but suppose we have a proposition about analyticity something like bachelor's or unmarried is an analytic truth is that itself analytic or synthetic that is to say when we describe something as analytic or as synthetic is that itself an analytic statement or is that a synthetic statement it's not that easy to say traditionally i think people thought of it as an analytic statement after all you understand the meaning of the term analytic its truth value is determined by the meanings of its terms and then you understand for example bachelor and unmarried and are and you put those together and you say yes it is true that in fact it's true by virtue of meanings not only that bachelors are unmarried but that it's analytic that bachelors are unmarried but quine is saying wait a minute that depends on a theory of meaning and a theory of meanings in empirical linguistic theory so actually we have to understand such classifications separating the analytic from the synthetic as themselves synthetic in other words one way of looking at his attack is to say it is a synthetic truth that bachelors are unmarried is analytic and if we think of it that way we realize hold on a second this whole thing is part of an empirical theory of the world so actually we can't separately from that take some sentences and put them aside and say they're just true by virtue of meanings or false by virtue of meanings whatever we end up thinking about that issue there's another way of thinking about analytic statements i've described them as things that are verbal truths but also as things that don't tell us anything about the world locke thinks of analytic truths as trifling propositions that is to say they give us no information about the world they convey no information at all except maybe about language like telling us what the word bachelor means and so they tell us nothing about the world as such well suppose we thought of an analytic statement as being defined that way not having its truth value determined by the meanings of its terms but instead as something that conveys no information about the world that you might say no empirical evidence could be relevant to it could not be confirmed or disconfirmed by anything about the world so you might say it's isolated from anything empirical so let's call an analytic statement understood in that sense empirically isolated if we understood things that way then we could say a synthetic statement is one that's not empirically isolated in other words it does give us information about the world so we could say look a synthetic statement if i learn that i learn something about the world but if i learn an analytic statement i learn nothing about the world all i'm learning is at best something about language if i think of things that way then i really start from a conception of what's empirical what is about the world in some sense what evidence could confirm or disconfirm and those are not exactly the same thing if i'm thinking about what conveys information about the world well is that the same thing as what could be confirmed or disconfirmed maybe but maybe there's things about the world that i couldn't confirm i couldn't disconfirm they are let's say facts about things outside my light cone i couldn't possibly get any information about them there could also be the classic empiricist cases of meaningless propositions that i ought to commit to the flames to use hume's phrase so if all that is right there actually two ways of understanding this idea of empirically isolated one would be to say that an analytic statement is empirically isolated in the sense that it conveys no information about the world at all another would be to say that it's something that experience could neither confirm nor disconfirm it would be a matter of well language perhaps or logic or something other than experience that would give us information about it so there are really two ways of understanding this the semantic one that says that an analytic statement conveys no information about the world at best about the uses of language whereas a synthetic statement conveys real information about the world beyond something about language but even that's tricky because after all if i describe the languages of the indigenous peoples of south america for example i'm doing it on the basis of empirical research i'm telling you things about language but on the other hand that's not analytic there is no analytic truth that tells me how many languages there are spoken in south america for example and so even then it seems like we're presupposing some notion of meaning to describe what constitutes information about the world or what constitutes information about language another way of looking at it is to say that empirical information could not give us any confirmation or disconfirmation but then we're in danger of grouping analytic statements with those that are just meaningless if i say something that is pure nonsense well that's not going to convey any information about the world and no empirical evidence will confirm or disconfirm it either but it's not because it's analytically true or analytically false it's just conveying no information about the world because it conveys no information at all so it is not very easy to mark out in this way exactly what analytic judgments are but let's take a step back as this distinction has been classically understood an analytic statement is true by virtue of the meanings of its terms and often we do have an intuitive grasp of meaning that tells us yes bachelors are unmarried that is something that is true by definition it's true by virtue of the meanings of the terms bachelor and unmarried we don't have to do empirical research we can't apply for a grant to study the question of whether bachelors are unmarried that's obviously foolish there is nothing to investigate it's determined to be true by the meanings of the terms there might be borderline cases cases where we have a hard time settling it like nothing can be both read all over in green all over or everything green is extended color terms are notoriously problematic and we might decide yes in some of those cases we really don't have a very good theory of meaning of the term that enables us to decide such questions so in practice i think most philosophers have said well even if quine is right that there are some borderline cases and even if it's true that in the end drawing the distinction is part of an empirical theory of language and meaning and so part of the science of semantics so it's part of the science of semantics that is to say it's partly the job of linguistics and to some extent the job of philosophers and logicians to figure out what the meanings of our terms are how to construct a theory of it and thereby what ends up counting as an analytic truth so it may be that calling something analytic or calling something synthetic is itself a synthetic judgment but that doesn't make it meaningless and it doesn't make us reject the distinction

Channel: Daniel Bonevac

Views: 2,057

Rating: 4.963964 out of 5

Keywords:

Id: 0XrV10tv_WQ

Length: 14min 6sec (846 seconds)

Published: Mon Oct 05 2020