Willard Van Orman Quine (1908–2000) was a prototypical American analytic philosopher. Following Frege, Russell, and Wittgenstein, he was concerned with how logic provides a foundation for mathematics, which in turn grounds physics and the other sciences.
We’ll be reading two of his most famous essays, both of which can be found in the collection, From a Logical Point of View
“On What There Is” (1948) discusses the ontological commitments of various theories, i.e. what entities these theories allege to exist. We know, for instance, that a word can refer to an object, but what if, as in the case of “Pegasus,” we know that there’s no such object? Frege said that the word still has a “sense,” and that moreover when two designations like “Superman” and “Clark Kent” refer to the same person, those terms still have different “senses.” Well, is the “sense” an object too? Frege is often accused of being a Platonist about senses and other abstract entities, meaning that while these don’t exist in space and time, they still exist in some sense, i.e. they have to be part of our ontology, our metaphysics.
Quine wants us to have as small an ontology as we need to support the sciences. While he doesn’t in this essay end up giving us an ontology, he does give us a formula for determining what ontological commitments a theory has. Just because we have a name Pegasus in our language doesn’t mean we’re committed to any creature picked out by that name, because (following Russell’s theory of definite descriptions) we can analyze a claim about Pegasus as saying “For all x, if x is a horse with wings, then such and such will be true of it.” There simply is no such x, yet we still understand the name, because we understand the description. The ontology of a theory is just the range of entities that “x” here draws from: “To be assumed as an entity is, purely and simply, to be reckoned as the value of a variable. In terms of the categories of traditional grammar, this amounts roughly to saying that to be is to be in the range of reference of a pronoun.” This might sound uninformative, but it provides a helpful way to think about theories: does my theory require that numbers, or universal concepts, or linguistic meanings, or quarks for that matter are real or not? Well, can I analyze statements made in the course of the theory such that I can omit explicit reference to those entities? This may require heavy-duty logical analysis.
In “Two Dogmas of Empiricism” (1951), Quine focuses on the notion of meaning. In keeping with his small-ontology philosophy (and an affection for behaviorism), he doesn’t like the division that thinkers since Kant had made between claims whose truth or falsity is a matter of the state of the world (“all bachelors are tidy”), and those made true (or not) by virtue of the meanings of words (“all bachelors are unmarried”); this is also known as the analytic/synthetic distinction. Quine thinks defining “analytic” is problematic: we can try to define it in terms of synonymy (words with the same meaning), but end up in a circle. Instead, Quine thinks that supposedly analytic claims really can be (rationally) changed in light of experience, and conversely that any claim can be held even in the face of counter-evidence. This is because our beliefs are a web, with some of them closer to the data of particular experiences and some of them (like the laws of logic, basic laws of science, and our definitions) closer to the center of that web.
The second “dogma” has to do with the behaviorist attempt to define meaning in terms of behavior, i.e. we can tell what a word means by how it is used, like what perceptual experiences it goes with. Despite his behaviorist suspicion of terms like “meaning” (which connotes some entity that he doesn’t want in his ontology, while “significance,” he thinks does not), he doesn’t think that this kind of reduction works, even in principle.
Buy the book
Both of these essays have been heavily anthologized and discussed in analytic philosophy circles. A famous response to “Two Dogmas” that you may want to look at is “In Defense of a Dogma,” by H.P. Grice and P.F. Strawson, which is outlined here. The thrust of their argument is that Quine is being too restrictive in what counts for an adequately clear term, and that we understand “meaning” enough to go on using it just fine.
Both of Quine’s essays can be seen as in dialogue with Rudolph Carnap, a student of Frege’s whose Aufbau
Carnap also responded to the attack on analyticity by challenging Quine’s claims about changing beliefs in light of evidence. The easiest way to get a handle on this is to read David Chalmers’s account of this; sections 1 and 2 of that essay are sufficient to give Carnap’s response, while the rest of the essay grapples with the positive challenge for Carnap then in distinguishing between cases where a belief in light of evidence really changes from cases where the belief stays the same but the definitions of the words involve change.
Listeners may also be interested in Quine’s own assessment of his “Two Dogmas” essay much later in his life (1991): “Two Dogmas in Retrospect.”
I can’t wait, I’m studying formal math and various logics, this is right up my ally.
It would be awesome if you guys did some “radical” philosopher like Rorty. Rorty doesn’t even think reference is a viable notion. The idea that words refer is questionable. No doubt people refer by their speech acts, by their intentionality, but words themselves strictly speaking do not. How could they? Talk of words referring is at best misleading and at worst dead wrong.
My two cents
Incidentally I think that point is basically shared by John Searle (I think I heard it on one of his Mind courses you can find on Youtube) and Chomsky has said something substantially similar (see the Stony Brooks interview, part 2 I think for the Philosophy one).
Glen
http://www.newappsblog.com/2012/10/on-the-origin-of-analytic-metaphysics-again-.html
From On What There Is Quine says this:
there is a gulf between meaning and naming even in the case of a singular term which is genuinely a name of an object.
To support this, he quotes an example from Frege:
The phrase ‘Evening Star’ names a certain large physical object of spherical form, which is hurtling through space some scores of millions of miles from here. The phrase ‘Morning Star’ names the same thing, as was probably first established by some observant Babylonian. But the two phrases cannot be regarded as having the same meaning; otherwise that Babylonian could have dispensed with his observations and contented himself with reflecting on the meanings of his words. The meanings, then, being different from one another, must be other than the named object, which is one and the same in both cases.
There seems to be seems to be an assumption in this that I don’t see has true. Maybe someone can clear it up for me. The assumption is that when the Evening Star was named that they were, in fact, naming a large physical object of spherical form, which is hurtling through space some scores of millions of miles from here. This is Quine’s name for related, but different, phenomena. It is not the phenomena which the Babylonian observer is naming. Quines naming does not include all of the phenomena that the Babylonian observer’s name includes, and vice verse. The phenomena that each namer is naming cannot be divorced from the particular relation the namer has with the phenomena, or, in other words, what each namer names is exactly what that phenomena means to him.
So, from this perspective, while it is true that Quine, the namer of the Evening Star, and the Babylonian namer of The Morning Star, all are meaning something different, they are not naming or referring to the same thing. Things or Objects is the name we have for the set of delineated existence. No object exists which has not been delineated as such by consciousness. Such delineation is the meaning of the object, and naming refers to that delineation.
This is the way I see it. I’m an amateur. Maybe someone can clear this up for me.
Hi Bill,
No expert on this, but I think the point that Quine might be making in this passage is a familiar one which (as you say) was originally made by Frege. The basic idea is the following:
– Take two co-referential terms: Reg Dwight, and Elton John, say.
– Observe that: if someone, S, believes that ‘Elton John is a good pianist’, say, he cannot derive – merely by reflecting on the content of his belief – the further belief that ‘Reg Dwight is a good pianist’ (if, on analogy with the Babylonian, he doesn’t know that Elton John and Reg Dwight are actually the same person).
– Thus, the meaning of some singular term – Elton John, the table, my foot – is not exhausted by what philosophers and linguists call its reference, i.e., the object it denotes.
I am not hugely familiar with the Quine paper, but think that it is the point he’s getting at in the passage you quote. Happy to be corrected though!
Does this address your question? What is the false assumption you think Quine’s making here?
http://spruce.flint.umich.edu/~simoncu/380/gricestraw.htm
this gives a 404 error in both Chrome and Firefox
Hi,
I’m just a layperson interested in philosophy with a question inspired by this podcast discussion on Quine that I hope someone can answer.
In the episode, and in the real world, it’s generally accepted that, as Lao Tzu said ages ago, “A cart is worth more than the sum of its parts.” In the show, examples are given, such as that the H2O molecules produce something very different than what they are separately. Consciousness from the assembly of biochemical parts is probably the most used example of this ’emergent phenomena’/’epiphenomena’ that illustrates WHY we can’t reduce (or expand) in certain ways. It’s also often considered to be problematic for causal determinism (and of course, logical positivism).
To play devil’s advocate (maybe as a kind of strict monist empiricist), why should the ‘novel’ quality of epiphenomena make a difference as to whether or not there is some consistent (causal) relationship in the behavior of combined elements in certain contexts? Shouldn’t it only matter whether the phenomena is a consistent behavioral result of the elements in a specific context, rather than that they are merely novel or unexpected phenomena? Doesn’t that consistent phenomena–even if we can’t explain it–just become ‘information’ at some point, just as the way that those things behave in that context? This is to ask, why can’t we say ‘As long as we consider that THIS IS WHAT HAPPENS TO THESE THINGS IN THIS CONTEXT WHILE WE”RE REDUCING/EXPANDING, we can justifiably reduce/expand’?
Consider what may be an oversimplified or loose analogy. Some people criticize consequentialism/utilitarianism based upon the thought experiment that even if we could somehow magically make everyone in the world happy by torturing a baby, we still wouldn’t do that because that’s not the kind of world we’d want to live in (Dostoevsky, Le Guin), so consequentialism isn’t tenable. But as some utilitarians note, we also need to INCLUDE THAT FACT (e.g. ‘torture isn’t compatible with any world we’d want to live in’) into the utilitarian cost/benefit analysis from the get-go. In a similar way, why don’t we just include the fact that removing or adding emergent phenomena is a fundamental part of the process of reduction/expansion, while still recognizing a deterministic framework?
I hope I conveyed my concern enough for someone to reply without too much effort. Thanks