Arnold Zwicky's Blog

A 2005 Savage Chickens cartoon (by Doug Savage) with what’s labeled as a “future perfect passive”:

The label isn’t exactly wrong — it alludes, somewhat indirectly, to the semantics of the material will have been disappointed with subject you and complement with your life — but the label invites comparison to material like amāverō ‘will have loved’ in Latin (expressing the “future perfect”). But English and Latin work very differently in how they package content into words.

When it occurs in a Latin sentence, amāverō is a single word, representing an inflectional form of the Latin verb lexeme AM- ‘love’; but when it occurs in in an English sentence, will have loved is a sequence of three words, each representing a form of an English verb lexeme (the “future” modal verb WILL in a FIN (“finite”) form; the “perfect” auxiliary verb HAVE in its BSE (“base”) form; and the main verb LOVE in its PSP (“past participle”) form. Latin packages the three bits of content in one verb word, English in three verb words.

In Latin, this packaging is done by assembling word parts, in a complex fashion; it’s all morphology. In English this packaging is a matter of pasting together three syntactic constructions, two involving a head V in combination with a VP complement:

(1) head: modal WILL  +  complement: VP with BSE head

(2) head: auxiliary HAVE  +  complement: VP with PSP head

and one involving a head V in combination with a variety of possibilities for complement: for instance, none:

(3a) head: main verb LOVE  (intransitive will have loved)

or a direct object:

(3b) head: main verb LOVE  +  complement: direct object NP  (transitive will have loved him)

If you know some Latin, it’s tempting to think of will have loved — or will have been disappointed or will have been being attacked— as a single word (and a regrettable number of sources on English structure talk this way). But Latin amāverō really is just one word (you can think of its parts as realizing, in a complex way, properties of the whole word), while English will have loved really is a sequence of three words, each with a syntactic life of its own: note, for instance, that the VPs involved can be syntactically complex:

will as a child have loved (will  +  as a child have loved)

will have passionately loved (will  +  [ have  +  passionately loved ] ])

That’s just a small sampling of the evidence in favor of one word for Latin vs. three words for English in this case. The larger point is that we can’t tell what’s going on in either language merely by thinking about the meanings of expressions; we need ways of arguing for one kind of analysis over another.

This entry was posted on November 26, 2012 at 10:07 pm and is filed under Constructions, Inflection, Linguistics in the comics, Syntax. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.