Two Dogmas of Empiricism

What are they, according to Quine?

1. The Supposed Analytic-Synthetic Distinction

2. Empirical Reductionism

Quine’s goals

“Modern empiricism has been conditioned in large part by two dogmas. One is a belief in some fundamental cleavage between truths which are analytic, or grounded in meanings independently of matters of fact and truths which are synthetic, or grounded in fact. The other dogma is reductionism: the belief that each meaningful statement is equivalent to some logical construct upon terms which refer to immediate experience… One effect of abandoning them is… a blurring of the supposed boundary between speculative metaphysics and natural science. Another effect is a shift toward pragmatism.

Note: Pragmatists call into question whether is it correct or helpful to talk about “belief,” and “meanings” for that matter, as something “in the head,” (my term, not theirs) and divorced from practice.

Quine is going to criticize view there exist “non-linguistic” meanings and/or “propositions.[1]” If he is right, these cannot be the right way to unpack “meaningful term,” “meaningful statements” or analyticity generally.

Quine will suggest that we must regard the meaningful sentence as deriving its meaning by the way the sentence is deployed within a language (in a way similar to the way that the meaning of a chess piece is the way it can be deployed according the of rules of chess and the context in which is appears on the board in the game). Scientific terms have their “meanings” by way of the practical role they play in the scientific project.

Quine Project in the Paper: Two Tactics

· Negative: trashing the “two dogmas” of logical positivism

o the analytic/synthetic distinction

o reductionism

· Positive: “naturalizing philosophy”

o Understanding ontology as a matter of pragmatic decision, comparable to the natural sciences

Quine’s plan

· Debunk the two central dogmas of Empiricism (“Logical Positivism”)

o He uses circularity arguments to show that we cannot come up with an adequate non-question-begging account of analyticity, without which the analytic/synthetic distinction cannot be maintained.

o Reductionism: “the belief that each meaningful [factual] statement is equivalent to some logical construct upon terms which refer to immediate experience.” Recall the provenance Locke et al. thought we should/ could perform on any meaningful term…trace back to simple sensations.

· “Naturalizing” philosophy (e.g. Turning formerly “philosophical/non-empirical questions into scientific questions)

o By “blurring the supposed boundary between speculative metaphysics and natural science” which was only sustained by the distinction between analytic and synthetic sentences, where the former are supposed to be the business of philosophy and the latter the business of the natural sciences.

The Circularity Arguments: Background for Analyticity

Quine Considers Historical and Popular Understanding of Analyticity

· A statement is analytic when it is true by virtue of “meanings” and independent of fact.

· Others suggest that statements are analytic when they are “necessarily true” statements.

An Analytic Statement is Characterized as…

· Hume: The denial of which is a logical contradiction.

· Kant: “one that attributes to its subject no more than is already conceptually contained in the subject.”

· Leibniz: “Truths of Reason” (as opposed to “Truths of Fact”) iff it is “true at all possible worlds.”

o A statement is contingent iff it is not necessary

Problems:

Kant’s definition

· restricts the definition to sentences in subject-predicate form

· it is metaphorical: what do we mean by “contained?”

Raises the questions:

· What is “meaning”?

· What is synonymy (sameness of meaning)?

· What are logical possibility and necessity?

· What is self-contradiction?

· What on earth are “possible worlds”?

Considers understanding Analyticity as a statement known to be true in virtue of meaning alone

Proposal: A sentence is analytic iff true in virtue of meanings and independent of fact.

Problem: What is MEANING?

The meaning of an expression (in the sense that interests us) cannot merely be its reference or extension.

· Extension is basically “referent”

· The extension of a singular term is the object it names. (“Venus” names the second planet from the Sun.)

· The extension of a general term, or predicate, is the set of objects of which the general term is true. (The “cat” is the set of cats.)

But two terms may have the same extension, but not “mean” the same thing.

· The extension of “Morning Star” is Venus.

· The extension of “Evening Star” is also Venus.

· Thus these two terms have the same referent/ extension, but NOT the same meaning[2].

· Meaning (whatever that might be) ≠ Extension

o Think like the Pragmatists here. At a minimum, competent language uses do not use the term “morning star” in the same way that they use the term “evening star.”

· The class of “Things with hearts is co-extensional with the class of things with lungs. But these classes so not mean the same thing. (EX: “Things with hearts” ≠ “Things with lungs”)

Quine:

“Once the theory of meaning is sharply separated from the theory of reference, it is a short step to recognizing as the business of a theory of meaning is simply the synonymy of linguistic forms and the analyticity of statements; meanings themselves, as obscure intermediary entities, may well be abandoned. (My emphasis)

Perhaps Meanings are “Intensions”

Problem: Analyticity has to be understood in terms of meanings (intensions) rather than extensions of terms

• But what are these meanings/intensions anyway?

• Properties? Individual concepts or Fregean senses (Gottlob Frege)?

According to Quine. We don’t need to talk about intentions or meaning, provided we can give an account of “synonymy” of linguistic forms which, in turn, will help give us an account of analyticity.

What are the linguistic rules for deeming to phrases as synonymous?

Rephrasing away “bad things”

Take the sentence:

Alice and Betty are of the same height.

· Surface grammar suggests that this attributes a 3-place relation to Alice, Betty and Height.

· “Height” seems to be a “thing” that Alice and Betty both possess.

But we can rephrase it as:

Alice is exactly as tall as Betty.

· What we really have is a 2-place relation between Alice and Betty

· Height has been exorcized!

o (i.e. we don’t need to account for some “thing” that each possesses.”)

Likewise, take the sentence:

“Bachelor” and “unmarried male” have the same meaning.

· Surface grammar again suggest that this says there’s a 3-place on “bachelor,” “unmarried male” and a “meaning” they both share.

· “Meaning” seems to be a “thing” that these phrases both possess.

But we can rephrase as:

“Bachelor” and “unmarried male” are synonymous.

· What we really have is a 2-place relation

· Meaning has been exorcized

o (i.e. we don’t need to account for some “thing” that each possesses.”

Great! So under what circumstances is it correct to say two terms are synonymous?

Synonymy

Even if there are no such things as “meanings” (Let’s assume that talk of meanings was a mistake.), we can make do with synonymy since we can use it to rephrase away “meanings” in the way that we paraphrased away heights.

If we can be successful here, we can try to define analyticity in terms of synonymy rather than “sameness of meaning.”

Let us instead understand synonymy as a property of words within a language. Given the rules that set up certain languages and the words they contain, certain words are deemed “synonymous,” which is simply to say that certain rules govern their use in reference to each other.

But now we have another problem: what is synonymy? What are these rules?

But. If we need synonymy to understand analyticity.

But how can we tell when terms are synonymous: Analyticity?

The problem of analyticity confronts us anew.

Two Types of Analyticity

1. Logical truths: (If we suppose a prior inventory of logical particles, comprising ‘no,’ ‘un-,’ ‘if,’ ‘then,’ ‘and,’ etc. then in general)… a logical truth is a statement which is true and remains true under all reinterpretations of its components other than the logical particles.

Ex: No unmarried man is married.

(No ~X is an X)

(Ax, Fx -> ~ (~F~x)

These are logical truths, and as such, are analytic tautologies in the least problematic sense.

2. The second “kind of analyticity” is what Quine is attacking here.

Ex: No bachelor is married.

Can this second kind of analyticity be turned into the first kind? (And thus be equally unproblematic.)

Yes, but only if we recognize “bachelor” and “unmarried male” are synonymous terms.

“The characteristic of such a statement is that it can be turned into a logical truth by putting synonyms for synonyms… (But) we still lack a proper characterization of this second class of analytic statements…inasmuch as we have had in the above description to lean on a notion of ‘synonymy’ which is no less in need of clarification than analyticity itself.

Thus, we need a rule for unpacking synonymy. Under what conditions is it correct to hold two terms as synonymous?

Perhaps:

Rule: X and Y are synonymous words iff swapping them is “allowed” within the language.

But…

Knowing that swapping “bachelor” for “unmarried man” is a licensed swap (in the language) requires already knowing that the terms are synonymous.

Therefore, this cannot very well be a TEST is synonymy.

Definitions: The Answer to Analyzing Analyticity? (Don’t count on it.)

Quine:

“There are those who find it soothing to say that the analytic statements of the second class reduce to those of the first class, the logical truths, by definition; ‘bachelor,’ for example, is defined as ‘unmarried man

But what do we mean by definition?

Every definition is either

1. Descriptive, i.e. lexicographical

2. Explicative

3. Stipulative

(Spoiler Alert: But none will do the job!)

Dilemma:

Descriptive Lexicographical Definitions (Dictionary Definitions)

· Bachelor = unmarried male that never has been married

· These are only descriptions of HOW we use and are “allowed’ to use the words in the current state of language, NOT prescriptive claims of how we must use the words.

· Thus definitions cannot/ do not MAKE two terms synonymous.

· We can ask whether dictionary definitions are correct, (i.e. whether they really capture synonymous terms).

· Since this is an open question, lexicographical definitions cannot unpack the notion of synonymy.

· And, of course, languages evolve. So, how people do in fact use the words is not a necessary feature of the words themselves, but rather on contingent fact based in current language practice. (Ex. “Gentleman” used to mean a male landowner.)

Explication Definition

• Explication (contextual definition):

“the purpose…is to improve upon the definiendum (term defined) by refining or supplementing its meaning. But even explication…does rest nevertheless on other pre-existing synonymies…[T]he purpose of explication is to preserve the usage of these favored contexts while sharpening the usage of other contexts.

Both types appeal to pre-existing synonymies

Lexicographical and explicative definitions assume that we already understand synonymy: for them “definition rests on synonymy rather than explaining it.” (My emphasis)

So we’re back to synonymy--and no wiser!

Stipulative Definitions

Stipulative definitions only go for a narrow range of cases

· Example: definition of logical operators in terms of primitives

· Example: ‘if p then q’ is defined as ‘~p or q’

These definitions aren’t arbitrary, but what fixes them depends on purposes in formulating artificial languages

· Thus, the most they can deliver is analyticity-in-L (e.g. In L, let A and B be related analytically.)

So we can’t appeal to “truth by definition” to understand analyticity.

How About Synonymy as Interchangeability Salva Veritate?

“A natural suggestion…is that the synonymy of two linguistic forms consists simply in their interchangeability in all contexts without change of truth value; interchangeability, in Leibniz’s phrase, salva veritate.

Can we understand synonymy as intersubstitutivity salve veritate (preserving truth value) so that, e.g.?

On this view:

Claiming that “bachelor” and “unmarried male” are synonymous comes to:

· For any sentence where “bachelor” occurs, “unmarried male” can be substituted and the truth value of the sentence will stay the same.

The question remains whether interchangeability salve veritate “…is a strong enough condition for synonymy, or whether…some non-synonymous expressions might be thus interchangeable.”

(i.e. thus interchangeability salve veritate might not be sufficient.)

So, even if necessary, it is not sufficient.

Point of Clarity: Cognitive Synonymy

“[W]e are not concerned here with synonymy in the sense of complete identity in psychological associations or poetic quality…We are concerned only with what may be called cognitive synonymy”

If “synonymy” is going to help us understand “analyticity,” then…

· What we need is an account of cognitive synonymy not presupposing analyticity.

· But determining interchangeability salva veritate is meaningless until relativized to a language

Quine Presents a Dilemma for this Approach:

Either the language is (1) extensional or (2) rich and non-extensional

1. In a purely extensional language, what we would regard as non-synonymous terms are intersubstitutable salva veritate (e.g. hearted organisms and kidneyed organisms, Morning Star and Evening Star)

· The morning star (Venus) rises in the morning. (True)

· The evening star (still Venus) rises in the morning. (Also True)

From the point of view of extension alone, it is true that the evening star rises in the morning, since this is true of what the term refers to (Venus).

2. However, these sentences are NOT intersubstitutable in our language because our language is NOT purely extensional. In a rich language, one rich enough to block the intersubstitutability of non-synonymous terms presupposes an understanding of analyticity

Thus: Extensional equivalence isn’t synonymy

Proof:

· “F and G are extensionally equivalent” means that all and only Fs are Gs

o Fact: All and only creatures with hearts are creatures with kidneys

· BUT “creature with a heart” and “creature with kidneys” are not cognitively synonymous

· Sam believes that fish are creatures with hearts – true

· Sam believes that fish are creatures with kidneys – false

· Sam understands the phrases “creatures with hearts” and “creatures with kidneys,” but knowing what these terms mean doesn’t help him here.

Therefore, extensional equivalence doesn’t capture the notion of cognitive synonymy.

Necessity as the Criterion for Analyticity or Synonymy?

OK.

But it is only a contingent fact that all and only creatures with hearts are creatures with kidneys. Perhaps if we strengthened the requirement from contingently coextensive sets to necessarily coextensive sets, it would work. (i.e. “Necessarily all and only Fs are Gs” does capture synonymy of F & G)

· This fixes the heart/kidney problem; while it happens to be the case that all and only creatures with hearts are creatures with kidneys, it is not the case that necessarily all and only creatures with hearts are creatures with kidneys.

And

· It is the case that necessarily all and only bachelors are unmarried men.

· If Tom understands what “bachelors” and “unmarried men” means (uses the term properly) then, he knows that Dick and Harry are bachelors if and only if he knows that Dick and Harry are unmarried men.

· So “bachelors” and “unmarried men” are cognitively synonymous

So far so good. But…

Quine claims we are not entitled to use “necessarily” here.

We could know that

1. “Necessarily all and only Fs are Gs”

ONLY IF we already know that

2. “Fs are Gs” is an analytic truth.

Remember: analyticity is what we were trying to explain but:

A) We tried to explain analyticity in terms of synonymy and then

B) explained synonymy of Fs and Gs in terms of a sentence using “necessarily”

C) But then it turned out that tacking “necessarily” in front of a sentence just says that that sentence is “analytic.”

So this account of analyticity in terms of necessity is circular!

Further: Intersubstitutivity is language relative

· The richer the language the less intersubstitutivity we get

o Example: Sentences of the form “If p then q” and “~p or q” are interchangeable in the truth-functional language of propositional logic. (Very un-rich language.)

· But not in ordinary English where “if p then q” says more.

o If today is Wednesday then it’s sunny.

Extensional Language Cannot Make Sense of Cognitive Synonymy

If this is true, they the whole discussion of “interchangeability” is a red herring/ waste of time.

Interchangeability won’t do. A language in which interchangeability is sufficient for synonymy must be richer than a purely extensional language. And the additional stuff will include words like “necessarily” which we can’t understand unless we already understand analyticity, which is what we were trying to understand in the first place.

Semantical Rules

Quine is alluding to Rudolf Carnap’s project which we will largely ignore.

The gist of Quine’s worry here is that appeal to semantical rules of a language, L, to explain analyticity only explains analyticity-in-L. But that doesn’t explain analyticity for any other language or, what we’re after, the notion of “analyticity as such.”

END PART I: The First Dogma

PART II: Empirical Reductionism

“[A] boundary between analytic and synthetic statements simply has not been drawn. That there is such a distinction to be drawn at all is an unempirical dogma of empiricists, a metaphysical article of faith.

We don’t know what analyticity is because we still don’t know what synonymy is.

Can we get at the idea of synonymy by appeal to the Verification Theory of meaning?

Verification Theory of Meaning & Reductionism

“The verification theory of meaning…is that the meaning of a statement is the method of empirically confirming or infirming it. An analytic statement is that limiting case which is confirmed no matter what. (emphasis added)

“Statement synonymy is said to be likeness of method of empirical confirmation or infirmation…[What] is the nature of the relationship between a statement and the experiences which contribute to or detract from its confirmation?...The most naïve view…is radical reductionism. Every meaningful statement is held to be translatable into a statement (true or false) about immediate experience.

If this works, we can understand both formal analyticity in terms of logical truth, and informal analyticity/cognitive synonymy in terms of likeness in verification/falsification conditions.

“So, if the verification theory (of meaning) can be accepted as an adequate account of statement synonymy, the notion of analyticity is saved after all.”

But for this to work the reduction process of synthetic statements to direct experiences has to be doable.

But is it? (Fat chance.)

Phenomenalism: An example of Reductionism

This raises the questions of what these methods of empirical confirmation are. What, in other words, is the nature of the relationship between a statement and the experiences which contribute to or detract from its confirmation?

“Locke and Hume held that every idea must either originate directly in sense experience or else be compounded of ideas thus originating…we might rephrase this doctrine in semantical jargon by saying that a term, to be significant at all, must be either a name of a sense datum or a compound of such names or an abbreviation of such a compound.

Two problems:

1. Physical Substance does not name a datum or compound of data.

2. This renders mysterious what “objects” are.

· Hume raises the specter of Phenomenalism as the only metaphysical position consistent with Empiricism and science. (See no “objects” as all.)

· Mill (and on some interpretations Berkeley) understood physical objects as “permanent possibilities of sensation.”

· Russell suggests they’re “logical constructions” out of objects of acquaintance—comparable to “the average plumber.”

“The first thing that appears when we begin to analyze our common knowledge is that some of it is derivative and some of it is primitive…the immediate facts perceived by sight or touch or hearing do not need to be proved by argument…[but] what is actually given in sense is much less than most people would naturally suppose and…what at first sight seems to be given is really inferred. (Russell)

“Objects of a kind, F, are said to be “logical constructions” out of objects of another kind, G, if the facts about Fs reduce to facts about Gs, or everything said using the F vocabulary may be said in a more basic or fundamental way referring only to Gs.

The Average Plumber is a Logical Construction.

· There is no such being as the Average Plumber over and above regular plumbers.

· The form of the sentence “the Average Plumber has 2.3 children” is misleading: “the Average Plumber” does not refer to any individual.

o (Rather a logical construction. What we mean is that if we divide the number of plumbers’ children by the number of plumbers we get 2.3)

Ayer reconstructs phenomenalism as the claim that talk about physical objects can be “reduced” to talk about experiences.

Verification Theory of Meaning and Synonymy

Russell's concept of definition, Quine claims, was an advance over the impossible term-by-term empiricism of Locke and Hume. The statement, rather than the term, came with Frege and Russell and to be recognized as the unit accountable to an empiricist critique.

“More reasonably, and without yet exceeding the limits of what I have called radical reductionism, we may take full statements as our significant units -- thus demanding that our statements as wholes be translatable into sense-datum language, but not that they be translatable term by term.”

Verificationists assume “that each statement, taken in isolation from its fellows, can admit of confirmation or information at all.”

So a Verificationist might claim that sentences are synonymous when exactly the same experiences count for and against them. And a sentence is analytic if all experiences count for it--in effect, if no experiences count for or against.

Quine’s Holism

But Quine rejects the idea that one can understand the meaning of a sentence purely in terms of its verification conditions. Instead he offers a Holism

“If we recognize with Pierce that the meaning of a sentence turns purely on what would count as evidence for its truth, and if we recognized with Duhem that theoretical sentences have their evidence not as a single sentences, but only as larger blocks of theory, then the indeterminacy of translation of theoretical sentences (or, mutatiis nutandis, meaning holism) is the natural conclusion.

“… Philosophers have rightly despaired of translating everything into observational and logico-mathematical terms. They have despaired of this even when they have not recognized, as the reason for this irreducibility, that the statements largely do not have their private bundles of empirical consequences.

“Our statements about the external world face the tribunal of sense experience not individually but only as a corporate body. There is no firm distinction between a linguistic and factual component of the truth of any individual statement. (Emphasis Added)

Empiricism Without the Dogmas

Some statements are more entrenched in our corporate body of beliefs than others, but all face the “tribunal of experience” together.

· So no statements are “purely linguistic” (Analytic)

· “The unit of empirical significance is the whole of science”

The Web of Belief

“The totality of our so-called knowledge or beliefs, from the most casual matters of geography and history to the profoundest laws of atomic physics or even of pure mathematics and logic, is a man-made fabric which impinges on experience only along the edges.

Thus, for Quine, knowledge and science is a field whose “boundary conditions” are experience.

· Every belief/ statement is, in principle, revisable.

· The difference between “logical principles” and empirical claims concerns the “costs of revision”

· The more central the belief within the overall ‘web,” the great will be the “cost” is revision.

· But these of course are “practical costs” to be measured against the practical benefits.

Consequences for Ontology

· Physical objects then, are merely “posits.”

· What we admit to our ontology is a matter of what we need for the best scientific theory.

o Physical objects?

o Sets? Sets of sets?

o Brick houses on Elm street?

o Protons?

o Ids, egos and super egos?

o Past lives?

So we are committed to the ontology needed by our best scientific theories where “best” theory is understood as “problem solving.”

Quine’s Radical Pragmatism

“Now Carnap ["Empiricism, semantics, and ontology," Revue internationale de philosophie 4 (1950), 20-40.] has maintained10a that this is a question not of matters of fact but of choosing a convenient language form, a convenient conceptual scheme or framework for science. With this I agree, but only on the proviso that the same be conceded regarding scientific hypotheses generally. Carnap has recognized that he is able to preserve a double standard for ontological questions and scientific hypotheses only by assuming an absolute distinction between the analytic and the synthetic; and I need not say again that this is a distinction which I reject.”

Likewise, when it comes to ontology everything is negotiable. As previously noted, ontological decisions are a matter of cost-benefit analysis reflecting on the needs of science which is merely an extension of common sense amd thus the ontological commitments of common sense.

But this serves to blur the distinction between philosophy/metaphysics and science. For both, it shifts questions of ontology in the direction of Pragmatism.

· The “correct” ontology is the one that best serves the purposes of the best science/ common sense.

· If science progresses (evolves) so that the best science requires a different ontology, then we adopt a different ontology.

My comment:

· If science progresses/evolves so that the best science requires a different ontology then we adopt a different ontology.

· But “progresses” and “evolves” are importantly different concepts. Strictly speaking there can be no progress since there doesn’t seem to be anything towards which we are “progressing” (or from which re are “regressing.”)

· Pragmatist C.S. Peirce talks about an “end of inquiry” towards which science is “progressing,” but it is hard to see how this isn’t just a poetic myth.

o If we get hit by an asteroid tomorrow, then the “end of inquiry” is today.

The Naturalization of Philosophy

“Ontological questions…are on a par with questions of natural science…the issue over there being classes seems more a question of convenient conceptual scheme; the issue over there being centaurs, or brick houses on Elm Street, seems more a question of fact.

“But I have been urging that this difference is only one of degree, and that it turns upon our vaguely pragmatic inclination to adjust one strand of the fabric of science rather than another in accommodating some particular recalcitrant experience.

“Conservatism figures in such choices, and so does the quest for simplicity…Each man is given a scientific heritage plus a continuing barrage of sensory stimulation; and the considerations which guide him in warping his scientific heritage to fit his continuing sensory promptings are, where rational, pragmatic.

---

[1] The difference between a “sentence” and a proposition is important here. If I said “Snow is white.” And then said again “Snow is white.” I have uttered two sentence. But one who holds to the existence or proposition would say I have expressed only one proposition. Likewise, were is to say “Schnee ist weiß.” Is have expressed the same proposition with a German sentence. On this view, the proposition is what the sentence is supposed to “mean.” But, as we shall see, Quine objects to this account of language and sentences.

[2] From the Pragmatist understand here, the meaning of the term is the role is plays within a language.