Theiling Online    Sitemap    Conlang Mailing List HQ   

Re: Test for middle voice?

From:Henrik Theiling <theiling@...>
Date:Sunday, November 20, 2005, 0:29

R A Brown <ray@...> writes:
>... In the strict sense, English does not have a middle _voice_, >i.e. a grammatically category distinguished morphologically from >active and passive. ...
I'd say that modern English does have some kind of systematic middle. More than than, say, German. You have a lot of transitive verbs that work well intransitively for the object* of the transitive verb used as the *subject* of the transitive one. This is quite close to a distinct syntactic category, and it is often kind of a middle voice: Verbs I have in mind: act: I fill the cup. med: The cup fills. in contrast to: pass: The cup is filled. Some things work in German, too, but less systematically (e.g. 'to cook' works the same in German, but 'to fill' needs a reflexive pronoun to work). Etc. The way I intuitively understood middle voice was looking at some examples of Ancient Greek. E.g. that 'to appear' (phainomai, the stem of 'phenomenon') is middle voice of 'to show'. It's clear that 'to be shown' (the passive voice) is not the same as 'to appear' (the middle voice) and this difference is just the point: it is shown, but no-one shows -- the agent is missing (nicely compares to 'the cup is filled' -- 'the cup fills)'. This example also shows (for me) quite clearly the link between the reflexive nature of the Greek middle voice, and OTOH, why many e.g. German uses a reflexive to immitate middle voice ('die Tasse füllt *sich*'): 'to show oneself' is close to 'to appear', and I think this reflexive nature stems from the 'pure' middle nature. Of course, an entity without any control cannot show (or fill) *itself*, but it appears (or 'fills'). This is the original middle nature. But when humans appear, they strictly show themselves (with control). This is a personal explanation, or way of understanding. But maybe it helps in understanding the middle voice intuitively. :-) **Henrik


Mark J. Reed <markjreed@...>