Why couldn’t Logic just be Descriptive?

Chris Matthew answered a few questions regarding logic and I wish to share them here. He was asked why logic couldn’t be descriptive and why couldn’t they just be in particular:

I think there are several problems here.
First, this “descriptive” approach to the laws of logic should be disambiguated further. It’s something that one hears often from village naturalists, but by itself, it tells us very little. Should we take this to mean that logical laws are little more than descriptions of how the human mind thinks (a la Quine), as would be plausible on some interpretations of Leibnizian rationalism? Should we take this to mean that logical laws are empirical generalisations about the fundamental structure of the universe? Should we take this to mean that logical laws describe public practices (e.g., Novaes 2015 or Sinan Dogramaci’s notion of “epistemic communism”)? Should we take this to mean that logical laws are built into the scaffolding of language in a way which reflects the world (a la Wittgenstein)? It’s not entirely clear.
Second, it appears to me that the question of whether to adopt a prescriptive or descriptive approach to logical laws (whatever that means!) covaries independently with the debate on Platonism, conceptualism, nominalism, and so forth. If the covenantal apologist avails himself of Anderson’s and Welty’s argument, for example, the fact that, say, the laws of logic refer to the structure of the world says nothing about the nature of the laws of logic themselves qua necessary propositions. The divine adverbial account of necessary propositions is entirely compatible with this description-oriented picture. Similarly, if the proponent of the descriptive thesis admits that logical laws are nonetheless necessary (as he must), the covenantal apologist may avail himself of an argument predicated on the modality of logical laws—and that without issue. Better still, if the covenantal apologist avails himself of an argument that rests on metaphysical queerness in the more general sense, there is no good reason to believe that one must be committed to the falsity of the descriptive thesis per se *or* nominalism for the argument to succeed.
Personally, I prefer to conceptualise the normative dimension of the laws of logic in terms of epistemic norms (as is recognised by the majority of contemporary philosophers). But inasmuch as the transcendental argument from logical laws (and the various sorts thereof) is distinct from the transcendental argument from epistemic norms (and I believe they are), our position would be none the weaker if it turns out that logical laws are entirely descriptive. …
Yeah, I’m not sure what “laws of logic [being] present in particulars” means. If it refers to the notion that, for example, “the law of identity is representative [instantiated?] in particular identities”, then this is trivially true—something that’s accepted by all parties to the debate. If it refers to the notion that there’s nothing universal about logical laws, then this undermines both the modal necessity of these laws and the connection that they bear to laws of thought—and thus it’s prima facie (radically) implausible.

One thought on “Why couldn’t Logic just be Descriptive?

Leave a comment