On Thu, Oct 8, 2009 at 5:14 PM, Spruiell, William C <[log in to unmask]> wrote:

Hey folks –

Weeks ago, we had an exchange on the definition of a sentence as “a complete thought,” and Beth asked if anyone had ever mentioned a source for that definition. I thought I remembered a good discussion of it somewhere, but I couldn’t recall exactly where. But now, I’m *supposed* to be working on program review documents (complete with mission statements), so of course I’ve suddenly remembered the reference I was looking for (and have thus provided perhaps the only extant example of a mission statement accomplishing anything useful).

It’s  Ian Michaels’s excellent _English Grammatical Categories_ (it focuses on English, as the title suggests, but he gives a detailed historical background dealing with the grammatical traditions that Renaissance England inherited). I’m doubtless oversimplifying the description Michaels provides (pp. 38-42), but in general the idea that “a sentence expresses a complete thought” appears to be one interpretation of a statement made by Dionysius Thrax, a Greek grammarian who died around 90 b.c.e. The idea was picked up by Priscian, a sixth-century Latin grammarian whose text was one of the core books used throughout the middle ages in Church schools (and in Europe, those were the only kind, really). With Thrax and even Priscian, though, “complete” can be construed as referring to whether a group of words accomplishes the speaker’s purpose, rather than whether it conforms to the more grammatically-based notion assumed in the modern definition of sentence. Medieval and Renaissance grammarians used several terms for groups of words – ‘oratio,’ ‘sententia’ –but none of these conformed strictly to those constructions that we’d call sentences, and no others. In some cases, ‘sententia’ could be roughly equivalent to ‘statement’.

I’ll venture a conjecture, which should not be taken as reflecting the views of Michaels (he doesn’t discuss developments post-1800): We developed a specialized sense of “the sentence” that both shaped and was shaped by punctuation patterns, and was keyed to the idea of a sentence being a full statement (a subordinate clause, by itself, has no truth value). But we kept using a translation of Priscian’s definition because….that had “always” been the definition (sorry, Brad).  And under one interpretation of Priscian’s definition, intentional fragments would be sentences.

Bill Spruiell

Dept. of English

Central Michigan University