The Brain, a User's Manual:
How to Define Things
By Gabriel Blanchard
The first step in all communication (and all reasoning is a form of communication) is just knowing what the words mean. Simple—right?
Closing Off Two Blind Alleys
Defining one’s terms is the first step in logic. In learning how ideas relate to each other, we first need to be able to fix a given idea in place. This is how unintentional ambiguities are avoided. The intentional ones have to be avoided by common honesty. The same is true of most fallacies, once we are aware that they are fallacies. One thing that logic will not do for us is the right thing. At the absolute most, it will tell us what the right thing is; choosing it is our business.
There is some circularity in any definition, which has to be accepted. For a native speaker of English to understand the meaning of a new word, it must be explained to them with—other English words. We rarely reflect on this difficulty, because most of us learned our native language long before we began assembling coherent, narrative memory. We therefore have a stock of words of all kinds that (in terms of conscious experience) we “come with” as intuitive, “default” vocabulary. Child language acquisition is a fascinating field in its own right, but not one the present author is competent to address. Luckily, it is also not one we need to fathom before we can learn how to define things.
What a Definition Needs
At this point, one might reach for a dictionary to see how it defines words. There’s nothing actively wrong with that—at least, not if one reaches for a reputable dictionary: the Oxford English Dictionary and the Merriam-Webster are good examples, and their recommended spellings align with British and American conventions respectively. (By contrast, the Super Dictionary is not a good example of a reputable dictionary, and may be something truly rare: a good example of nothing whatsoever.1 But we digress.) However, learning how to define by reading a dictionary extensively enough to imitate it would be an imperfect, time-consuming process.
Using the following principles is more practical. To help illuminate them, we shall consider the definition of the word “chair”—we shall consider it; whether we shall achieve it is another question.
The White Rabbit put on his spectacles. "Where shall I begin, please Your Majesty?" he asked.
"Begin at the beginning," the King said gravely, "and go on till you come to the end: then stop."Lewis Carroll, Alice's Adventures in Wonderland, Ch. XIII: Alice's Evidence
i. A definition must cover the entirety of one sense of a word. There is no reason to tie our brains in knots figuring out how to define “chair” in a way that will somehow cover both the thing that sits next to the kitchen table and the leading position of an organization. Those are both “chairs,” yes, but in different senses; we may legitimately distinguish one sense of a word from another. However, when we speak about the group of things we need to fill an auditorium with before the principal addresses the school, that is a chair in the same sense as the thing by the kitchen table; our definition needs to encompass both, along with everything else that is “that sort” of chair.
ii. A definition must exclude those things which are not meant by the word in question. According to a story about the philosopher Diogenes, he once overheard Plato define man to his students, a little flippantly, as “a featherless biped.” Diogenes went away, got a chicken, plucked it, and came back, brandishing it before the Academy and saying, “Behold Plato’s man!” The gentlemen of the Academy responded by adding “having broad, flat nails” to the definition of man (which does not seem to have provoked further sarcasm). In our definition of “chair,” we want to exclude not only thoroughly unrelated furniture—dressers, bookshelves, stationary bicycles, etc.—but also other forms of seating, like sofas or ottomans.2
iii. A definition should focus on relevant or essential information, not incidental information. The now conventional philosophic definition of man, which comes to us from Aristotle, is “a rational animal.” Man is naturally featherless, and does naturally possess broad, flat nails, so the Platonic “featherless biped with broad, flat nails” can’t exactly be called untrue, but these things are not what distinguishes man from other kinds of being (e.g., they are also true of a few other primates). Returning to “chair,” every chair will necessarily be of some particular color, or set of colors, but for that very reason, listing all the colors a chair could be is not really part of what the word means. We want to know from our definition what only and always makes something a chair; put another way, we want to know the essence of chair-ness.
iv. A definition should not be unduly informal. This rule mostly comes in to cure students of the habit of giving definitions like “X is when” or “X is where.” Unless time or place actually enter into the meaning of a term, the words “when” and “where” should not do so, to avoid confusion.
v. If at all possible, definitions should be impartial. The present author selected the word “chair” for this post, after deciding against a different exemplar for definitions, namely “sandwich,” a term well-known for attracting controversy. Obviously a great deal of this controversy is jocular; yet, joking or not, it still puts the definer in the position of having to decide whether the quesadilla, the taco, the hamburger, the frankfurter, the wrap, or even—counterintuitively—the open-faced and bagel sandwiches, are sandwiches.3 “Chair” admits of some disagreement, but far less; evidently chairs excite human passions less than sandwiches do.
An Attempt
Alright. With these five rules in mind, let us try to define our chosen word.
Chair (noun: plural, chairs). An object designed for one person to sit on, with a back and at least one leg.
This does seem to get in what makes a chair (deliberately made, for the purpose of sitting, carrying capacity) while excluding things that are not chairs (e.g., “with a back” distinguishes chairs from stools). Let us compare this effort with the dictionaries mentioned above.
A seat for one person, typically movable, supported by four legs or feet, and having a rest for the back. —The OED
A seat typically having four legs and a back for one person. —Merriam-Webster
We appear, if anything, to be outstripping these luminaries! Both have forgotten rule iii, and thus been put in the ludicrous position of implying that a three-legged chair is not a chair; the Merriam-Webster can at least plead that it has couched (heh) the extraneous property of leg numbering in the reservation “typically,” but it did not need to cite it at all.
A Precaution
It is, of course, great fun to catch out one’s betters, elders, or opponents in a shoddy definition (whether of mere chairs or of something as important as sandwiches). However, it must also be remembered that the meanings of words change with time—not usually very quickly, but perceptibly—and nearly every word, in any language, will have more than one sense. Definition does not exist to let us play “gotcha”; it exists so that we can clarify our thoughts, both for the sake of sharing them and for the sake of understanding others. The point is, all definitions are provisional and exist for a purpose: to convey meaning, thought, truth, between one mind and another. Truth is a cooperative pursuit, not a zero-sum game; if my sworn enemy discovers something true and tells me about it, we both win. It is perilously easy to let self-conceit, rivalry, pettiness, and suspicion corrupt and destroy the pursuit of truth. We must decide, at every moment the choice presents itself, to prefer truth to ego.
1Those who have seen this image have, wittingly or not, been exposed before now to the stylings of the Super Dictionary; we refrain from adducing further exemplars of its “definitions,” as longitudinal studies on its general safety for developing minds have yet to be conducted.
2Sorry, mom, we know you said not to use it as a seat—but it’s just so convenient.
3In order, the answers are of course no, yes, yes, no, yes, no, and no (the quesadilla is essentially a type of pie; the taco is a sandwich for the same reason as the hoagie; the hamburger is a sandwich in the classic sense; the frankfurter is sui generis; the wrap is in substance a horizontally elongated, folded-over taco; the “open-faced sandwich” has no between-ness and is therefore ineligible, unless folded in half; and the bagel “sandwich” isn’t on bread, it’s on a bagel, QED. (Burritos remain an open question, though they may essentially be a kind of beef Wellington.)
Gabriel Blanchard is CLT’s editor at large. He lives in Baltimore, MD.
If you liked this piece, check out our earlier posts on fallacies like circular reasoning, the use of straw men, failures of epistemic humility, dishonest appeals to the First Amendment, and selective treatment of evidence. Or perhaps you’d like to change tacks for a bit and look through the catalogue of our Author Bank—it includes dozens of writers, from Hippocrates, Procopius, Magna Carta, and Christine de Pizan to Adam Smith, Hans Christian Andersen, and Willa Cather. And thank you for reading the Journal.
Published on 10th October, 2024. Page image of John Tenniel’s illustration of the Court of Hearts in Alice’s Adventures in Wonderland, ca. 1890.