The Scope of Semantics

(AKA Katz’s Semantic Theory (Part II). This post discusses chapter 1 of Jerrold Katz’s 1972 opus. For my discussion of the preface, go here.)

If you’ve taken a semantics course in the past decade or two, or read an introductory textbook on the topic published in that time span, you probably encountered, likely at the outset, the question What is meaning? followed almost immediately with a fairly pat answer. In my experience, the answer given to that question was reference1—the meaning of an expression, say dog, is the set of things in the world that that expression refers to, the set of all dogs in this case. Now, I can’t exactly recall my reaction the first time a teacher presented that as an answer to the question of meaning. I might have been wholly unimpressed, or I might have had my mind blown, that way that an impressionable young mind can be blown by someone giving a pat, confident answer to a deep question. Either way, I know that every time I’ve heard that answer2 to the question of meaning since, it’s become less impressive, to the point of being slightly offensive. At best, a pat answer is incomplete; at worst, it’s flat wrong.

Of course, I never really had a better answer to the question of meaning, and most of the other answers on offer seemed much worse. I couldn’t shake the unease I had with reference as an answer, but I couldn’t fully articulate that unease. Which is why I was very quickly drawn into Semantic Theory—Katz pinpoints and articulates the source of that unease on page 3 of the book:

The misconception, it seems to me, lies in the supposition that the question “What is meaning” can be answered in a direct and straightforward way. The question is generally treated as if it were on par with questions like “What is the capital of France?” to which the direct and straightforward answer “Paris” can be given. It is supposed that an answer can be given of the form “Meaning is this or that.” But the question “What is meaning?” does not admit of a direct “this or that” answer; it’s answers is instead a whole theory [emphasis added]. It is not a question like “What is the capital of France?” “When did Einstein retire?” “Where is Tasmania?”because it is not merely a request for an isolated fact, a request which can be answered simply and directly. Rather it is a theoretical question, like “What is matter?” “What is electricity?” “What is light?”

(Katz 1972, p3)

Imagine if, instead of developing theories of matter, electricity, and light, the early physicists had been satisfied with giving a simple answer like Matter is anything you can touch and feel. We wouldn’t have a science of physics, or chemistry. We likely wouldn’t have any science as we know it.

Katz goes on to acknowledge that, if one were to ask a physicist what electricity is, they might give a simple answer, but notes that such an answer would be a highly condensed version of the theory of electromagnetism that has been developed over centuries of inquiry. Similarly, if you were to ask a phonologist what a syllable is, or what pronunciation is, or if you asked a syntactician what a sentence is, or what grammar is, you might get a similar condensed answer with a several big caveats. You certainly wouldn’t get a simple straightforward answer. In fact, one of the first tasks in any introduction to linguistics is to disabuse students of any simple answers that they may have internalized, and even to disabuse them of the notion that simple answers to such questions even exist.

This seems to leave us in a bit of a bind. If we don’t know what meaning is, how can we study it? Katz’s response: the same way we did with chemistry, biology, phonology, etc.—We identify a set of phenomena that are definitely under the umbrella of meaning, and go from there. Not to disappoint, Katz identifies 15 such phenomena which he frames as subquestions to the meaning question:

  1. What are synonymy and paraphrase?
  2. What are semantic similarity and semantic difference?
  3. What is antynomy?
  4. What is superordination?
  5. What are meaningfulness and semantic anomaly?
  6. What is semantic ambiguity?
  7. What is semantic redundancy?
  8. What is semantic truth (analyticity, metalinguistic truth, etc.)?
  9. What is semantic falsehood (contradiction, metalinguistic falsehood, etc.)?
  10. What is semantically undetermined truth or falsehood (e.g., syntheticity)?
  11. What is inconsistency?
  12. What is entailment?
  13. What is presupposition?
  14. What is a possible answer to a question?
  15. What is a self-answered question?

A formidable list to be sure, but, as far as I can tell, modern formal semantics only cares about 11–143. Katz expands on each of these with representative examples. I won’t go into those examples, but they all are based on intuitions that a person would have about linguistic meaning. If one takes these as the leading questions of semantic theory, Katz argues, then the simple answers to the meaning question lose their appeal, as they do not answer the subquestions 1–15, or at least cannot do so without a complex semantic theory to supplement them.

Furthermore, Katz points out that the debates between the competing simple answers all use arguments based on the phenomena that 1–15 as about. Take, for instance, the best known critique of the referentialist answer. If we assume that meaning=reference, then any two expressions that have the same referent, must be synonymous. Gottlob Frege, the godfather of formal semantics, argued that there were expressions which had different meanings but had the same referent, the classic example of which is the morning star and the evening star. The two expressions have different meanings (they differ as to when the star appears in the sky), however they refer to the same object (the planet Venus). And once you start to think about it you can come up with a seeming infinity of such examples.

Katz goes on to show that critiques of other simple answers to the meaning question are based on what hes call “strong pretheoretical intuitions,” all of which raise at least one of questions 1–15. His point here seems to be that we can’t divorce our semantic theory from pretheoretical intuitions such as the ones the form the basis of 1–15, so why not just embrace it? Why not throw away the “leading principles” and just try to build a theory that answers 1–15?

Katz closes the chapter by discussing skepticism with regards to meaning. It’s hard to honestly maintain skepticism, he argues, when we can marshal an extensive body of evidence that meaning exists. That body of evidence starts with an explication of 1–15, but likely extends beyond that. It is even harder to honestly maintain skepticism if we can build a theory that shows the regular and law-like behaviour of the evidence marshaled. Taking a suggestion from Quine (who played a major role in the preface), Katz compares the situation that he finds himself in to that which ancient astronomers found themselves in:

Astronomy found its answer to “What are planets?” by constructing a theory that explained planetary motion on the assumption that planets are physical objects that obey standard mechanical laws. In the same spirit, once we construct a theory that can successfully explain a reasonably large portion of semantic phenomena, we can base our answer to “What is meaning?” on what the theory had to assume meaning was in order to provide its explanations.

(Katz 1972, p10

Semantics, as it is taught and studied today, is commonly considered by non-semanticists to be the most arcane and opaque subfield of linguistics. It’s not clear what is more obscure, the questions that semanticists ask or the formalism that they use to answer those questions. I often wonder if there is something endemic to questions of meaning that make them seem arcane to many, or if it is a failing in the standard answer that leads to this feeling. This chapter of Katz’s book, for me, rules out the former. The questions in 1–15 are far from arcane, or, at least, they’re no more arcane than the questions that occupy the other subfields of linguistics. Maybe if we took Katz’s view of semantics, fewer students would run screaming (or grumbling, or yawning) from semantics classes.

In the next chapter, entitled “On the general character of semantic theory” Katz begins constructing his theory.


Footnotes (the links might not work, sorry)

  1. ^ I learned my semantics in a generative department where reference was the answer. Other departments might have had another answer.
  2. ^ and sometimes I’ve even given that answer as a teacher.
  3. ^ Entailment and inconsistency are the key phenomena. Presuppositions are useful as diagnostics. Questions have only recently gained currency lately it seems.

One thought on “The Scope of Semantics

Comments are closed.