← Return home

Chicken/Egg

William Dempsey

Recently, during a social dinner, a friend said to me that he thought the chicken/egg riddle was uninteresting, because (to paraphrase) “it was obviously the egg that came first.” I disagreed with the former claim: I do not think the riddle is uninteresting, though I do think there is a sense in which it might be understood as being trivial, and thus not a very good riddle, if we are to understand riddles as kinds of things that demand some real effort to uncoil. 

In terms of its immediate effects, the chicken/egg is more like a thought experiment, or an “intuition pump”.[1] Introduce it, and expect relatively quick (and not an insignificant amount of the time confident) interlocutor response. “Obviously, it’s the egg”; “Obviously, it’s the chicken.” Optionally followed by (at least seemingly) genuine incomprehension that someone else might endorse the opposite answer. This gives one clue about what makes the riddle interesting.

To be a little more precise, I am aware of three common reaction schemas:

(1) Confident, determinate response that it is obviously one and not the other.

(2) Unconfident but determinate endorsement of one over the other.

(3) Claim that it is genuinely indeterminate.

I want to add the following two to the list, though in my experience they are far less common:

(4) Claim that it is genuinely paradoxical.

(5) Claim that it is ambiguous, and that the answer is meaning-relative.

(5) is the response I wish to endorse as correct, but in doing so I mean to show how said view helps to explain the putative plausibility of (1)-(4).

I’ll start with a further explication of (5). The basic assumption of semantics is that linguistic items (words, sentences, etc.) have meanings. Meanings are constantly appealed to in conversation and discourse, but also remain an extremely peculiar philosophical object. One way of viewing 20th century philosophy as a whole—on both sides of the analytic/continental divide—is as a period in philosophy foundationally preoccupied with semantics (thus with meanings), rather than, say, metaphysics, or epistemology, which are often associated with earlier epochs of philosophy.

The notion of ‘meaning’ has been, over the past century, systematically distinguished from other linguistic features and properties. Semantics is thought of as distinct from syntax, and, within semantics, the work of especially Frege, Church, and Carnap, among others, has generated what was supposed to be a useful distinction between extensional meanings and intensional meanings for words. However, as Putnam was to point out,

The canonical explanation of the notions “intension” and “extension” is very much like “in one sense ‘meaning’ means extension and in the other sense ‘meaning’ means meaning. […] It is as if someone explained the notion “probability” by saying: “in one sense ‘probability’ means frequency, and in the other sense it means propensity.” “Probability” never means ‘frequency’, and “propensity” is at least as unclear as “probability.”[2]

It is obviously not my intention to try to work out an adequate theory of meanings here, but I hope that the notion is clear enough to be understood intuitively, even though we are still very far from a complete explanation. Anyone familiar with Frege will be familiar with his notion of ‘sense’[3]—senses are just intensions, that is, meanings. Even those who have not read “Sense and Reference” are likely to be familiar with the language of sense. The basic idea is that many—possibly all—words possess multiple senses. Hence we can say “in one sense, to be green means to be a certain primary color that stands in certain determinate relations to the other primary colors red and blue; in another sense, to be green means to be a novice.” Accordingly, many or all words are ambiguous, insofar as they mean different things in different contexts. Occasionally, one finds herself in a situation in which she is unable to determine which of multiple candidate meanings is the meaning actually in use for a given word.

The other major part of Frege’s apparatus is his descriptivism. According to descriptivism, meanings of words are descriptions associated with those words.[4] Think of the idea of a definition. On this view, what a word means is the same thing as how that word is defined, which is to say, what other words we use to explain said word.[5]

We can understand reaction (5), then, as suggesting that there is some ambiguity in the meanings of at least some of the words in the question “Which came first, the chicken or the egg?”. The crucial ambiguity, I think, is in the word ‘egg’. If ‘egg’ is taken to mean ‘egg from which a chicken came’, then the answer to the riddle is that the egg came first. In fact, this answer looks to be analytic. Or, if ‘egg’ is defined as ‘chicken egg’, as in, ‘egg laid by a chicken’, then the answer is evidently that the chicken existed before the egg. The question turns on whether we define ‘egg’ (that is, ‘chicken egg’) as denoting an efficient or final cause of the existence of a chicken. This is what I mean when I say that the riddle’s answer is meaning-relative.

The meaning-relativity hypothesis—(5)—then permits the explanation of why (4) could seem true. If, relative to the sense of ‘egg’, (disjunctively) both the propositions ‘The chicken came before the egg’ and ‘The egg came before the chicken’ are true, then, for someone who does not recognize the first part of the preceding fact—that the propositions are disjunctively (in fact, this is an exclusive disjunction) true—it will seem to her that the riddle’s output is a genuine contradiction, a paradox. With the Fregean apparatus introduced, however, this seeming paradox can be revealed to be nothing more than a case of ambiguity.

The same methodology works to dissolve the worry that the answer to the riddle is indeterminate, as in (3). Indeterminacy and paradox are really two sides of the same coin, depending on what route one decides to pursue once one encounters a case of apparently renegade logical status (the most conspicuous signpost that you’ve reached this territory is, of course, the liar paradox). 

We are also now in a position to give at least a partial account of reactions (1) and (2); in both cases (which differ only by the relevant cognizer’s confidence in the correctness of her opinion), the person who asserts that one answer is obviously right is operating according to a conceptual framework on which ‘egg’ is defined a certain way, and does not recognize the possibility of an alternative framework on which the meaning of ‘egg’ differs.

A related remark on differing conceptual schemes:  at dinner, my own interlocutor had suggested that perhaps there could have been no chicken/egg riddle before the advent of evolutionary theory. I do not think this has to be true, though I don’t know anything about the actual history of the riddle. In a pre-Darwinistic, theological-creationist framework for biology, one can still ask whether the chicken or egg came first, but this amounts to a different question than it would in the context of evolutionary theory; on the theological worldview, the chicken/egg riddle is a question about God’s preferences, or His personality—the question becomes: would God decide to create eggs, that would hatch into chickens, first, or would He begin with the chickens themselves?

The chicken/egg is not so much interesting for the question(s) it asks. Rather, it is interesting, I think, ultimately in that it is a nonparadox masquerading as a paradox, and a semantic plurality masquerading as a unity. It is interesting for the conditions it presupposes in order to be sensical. It is also interesting for the way we interact with and respond to it, and for the metaconsternation it can elicit insofar as we may recognize, as my interlocutor did, that it really should be a trivial riddle. Yet, for various reasons, it does not unreservedly come across as one.

[1] Daniel Dennett’s phrase.

[2] Hilary Putnam. “The Meaning of ‘Meaning’.”

[3] Gottlob Frege. “Sinn und Bedeutung.”

[4] Fregean descriptivism was famously attacked in Saul Kripke, Naming and Necessity.

[5] This is, of course, not the only live theory of meaning, in addition to Kripke’s causal theory (see note 4), another recent candidate of some significance is Brandom’s inferentialism, which is developed in Robert Brandom, Making it Explicit.