Friday, March 06, 2009


Life is full of annoying things, and for most of us it is too short to bother with Greek accents (or at least, that's what we tell ourselves). So let me confess my latest act of utter despair, I had a look at the phrase πρός σε. Σύ is one of those words that has accented and unaccented forms. In its unaccented form (only available in Gen., Dat., Acc.) σύ throws its accent back on the previous word; it is an enclitic. According to Moulton II.180 the enclitic forms of the personal pronouns ἐγώ and σύ 'are not used after prepositions, except with πρός (generally).' This is confirmed (well, repeated at least) by BDF par. 279.

I searched my Bibleworks for the phrase in the NT (19 hits) and checked each of them against my trusted, printed NA27 (no differences).
Here is the result: πρός σε in the first two instances (Mat 14:28, 25:39), in all remaining 17 cases both words are accented.
Is there an explanation (or even a MS tradition!) or are we, in the words of the Dutch painter Karel Appel, 'just messing about'?


  1. I think the 'rule' in this case is that it's wholly a matter of editorial emphasis.


  2. Do you want us to guess what went on behind the guess work of Greek accent marks? This is a question on theory if you ask me.

  3. Dirk,

    Isn't it the case that both forms are possible, one unemphatic and the other emphatic? So the MS tradition mostly seems to be recording an emphatic σέ rather like choosing the emphatic εμέ over the enclitic and unaccented με. [sorry - cannot find the breathing at the moment!]


  4. Yes, both the emphatic and enclitic forms are perfectly acceptable. I am just wondering if the praxis of NA27 is based on an - unexpressed - editorial policy or whether the choice for either form goes indeed back to a manuscript tradition.

    There seems to be no inconsistency within NA27 regarding πρός με, but only with πρὸς σέ. It is time that someone has a good look at scribal practice in some minuscules.

  5. I want to thank Dirk for this. The accents are very important and there are still little mysteries to unravel.

    How would we know how to read Greek if it weren't for the accents?

    On the otherhand, the accents in our texts are mostly 'blind', according to rule and not according to the communicative context. I hope to present a paper in the 'near' future, within a couple of years,that will demonstrate the artificial nature of the accute/grave flip. This is somewhat related to Dirk's observation, and I would add that pragmatically the 'grave' accent really means 'optional high tone', use a high tone "if/when you need it". This has especially come into focus as I have been recording long stretches of GNT.