2.6 Definitions

Category: Logic and Mathematics

Keywords: definitions, symbol, symbols, definition, meanings, meaning, term, meaningful, symbolic, define, defined, meaningless, operations, implicit, usage

Number of Articles: 316
Percentage of Total: 1%
Rank: 44th

Weighted Number of Articles: 532.9
Percentage of Total: 1.7%
Rank: 9th

Mean Publication Year: 1950.1
Weighted Mean Publication Year: 1957
Median Publication Year: 1950
Modal Publication Year: 1934

Topic with Most Overlap: Meaning and Use (0.0442)
Topic this Overlaps Most With: Analytic/Synthetic (0.0437)
Topic with Least Overlap: Duties (0.00047)
Topic this Overlaps Least With: Population Ethics (0.00286)

Definitions

Figure 2.23: Definitions

Definitions Articles in Each Journal

Figure 2.24: Definitions Articles in Each Journal

Comments

At first when I saw this topic come up, I assumed that it would be part of the influence of positivism. If positivism is going to work, you’re going to need a lot of terms to be defined, and then you’ll want a theory of definitions. And the timeline, especially on the single graph, checks out for that explanation. But I’m not sure that’s all of what is going on. If you look through the notable articles that are clearly about definitions, such as the Dubs or Reid articles mentioned above, they don’t seem super positivist. They don’t seem openly hostile to positivism, but they don’t motivate the search for a theory of definitions with an account of their role in positivist philosophy.

That said, they also don’t come up with things that would pass muster by contemporary standards as theories. I’m going to quote Dubs’s conclusion at length because I can’t really summarise it otherwise.

The foregoing summarizes the important features of what I find necessary for definition. To recapitulate: definitions are of various types, depending upon what function they are to fulfil. Dictionary definitions attempt to explain a meaning to a person who is ignorant of it. The three requirements for such a definition are commensurateness, understandability, and (usually) concept- uality. Scientific definitions attempt to enable the exact identification of cases of a term in experience or thought. There are two requirements: commensurateness and definition only in terms previously defined. Such definitions are conceptual or non-conceptual. Conceptual definitions, in any developed science, form an elaborate hierarchy, the creation of which is sometimes an extremely difficult task. Non-conceptual definitions may be causal definitions, denotative definitions, or semi-denotative definitions. In a few sciences, essential definitions have been achieved; most definitions are merely nominal. I hope that the foregoing account of definition fits the varied facts of scientific and logical usage and constitutes an intelligible and consistent account of definition. (577)

I mean, sure definitions are conceptual or non-conceptual; that doesn’t seem like a stunning conclusion. And more generally what we get here feels more like a shopping list than a sharp account. But it’s still something, and Dubs does make some good points along the way about the mischief that philosophers can do with badly drawn definitions.

One other thing to note about this topic is that it is very American. The articles here are primarily by American authors in American journals. And for all my complaints about Dubs, there is a kind of professionalism to the American writers that isn’t always shared by their British contemporaries. This professionalism will be, I’m sure, not always admired by those who look to philosophy for deep insight. But there is a sense here, in a way that there isn’t in the earlier topics, that we’re getting early versions of philosophy as it is now done.

This topic is helpful for seeing the difference between the raw counts and the weighted count. The weighted count is much higher, as you can see above. This is because every time someone introduces a definition in their paper (as such) the model gets a bit excited and thinks that it is a paper on definitions. It quickly figures out that this probably isn’t the case, but it can leave a residual probability. Those residuals add up, and eventually you get that the topic is the 9th highest by weight out of the 90. And that’s why even though there are very few papers in these journals that are in any sense on definitions since the 1980s, the graph doesn’t go to 0. All those times the model thinks it is 2 or 3 percent likely that a paper is about definitions add up, and give you a graph like this one.