From: Daniel Ria–o (
Date: Thu Jun 03 1999 - 11:59:28 EDT

Carl W. Conrad wrote:

> The
>problem is how to make sense of the dative THi ELEUQERIAi with which the
>clause begins. Brian originally asked whether this is a Dative of Advantage
>or a Dative of Purpose. I originally said that the noun in a Dative of
>Advantage construction should normally refer to a person, but "Freedom"
>doesn't seem to be personified here. As for a dative of purpose, as I said
>in my original reply, although this is a very important function for the
>Latin dative, I really don't know of a Dative of Purpose in ancient Greek,
>and before I'm ready to accept that, I would really like to see some
>convincing instances of it in Hellenistic or earlier Greek.

        This is a very interesting question. I can't deal in extenso now
about all the theoretical implications, but maybe a short summary can be of
        First of all, a mass of evidence argues for the presence of the so
called Dativus finalis (= DF, or Dative of purpose) not only Latin, but
also Ancient Indian, Avestan and Church Slavonic. If we do not suppose that
the construction is of Indo-European Origin, then the onus probandi belongs
to whoever tries to demonstrate de apparition of the construction
independently in every of those languages. But I also think that Micenic
has incontrovertible examples of dativus finalis, quoted in Vilborg's
tentative Grammar of Micenean (which I don't have at hand).
        On the other hand, there are dozens of adnominal datives with a
clear "purpose" meaning in the papyri from at least s. III a.C. on. Mayser,
Edwin, *Grammatik der griechischen Papyri aus der PtolemŠerzeit.*
Berlin-Leipzig 1926-34, Vol. 2.2.1 p. 146 (¤ 74.2), without trying to be
exhaustive, quotes many examples like the following:
        a)nh/lwma e)rga/tais *PZen.*59013.1 (III a.C.)
        u)povmnhma para\ Xe/nwnos Mi/lwni *PEleph.*17.1 (III a.C.)
        ei)s trofh\n toi=s pw/lois *PSI* 6.584.26 (III a.C.)

The purpose construction is also found in Attic inscriptions from the IV
century a.C. on, like:
        csu/la kai\ a)/nqrakes to=i molu/bdo=i (with "o" for "w", "cs" for
"x") IG 1[[2]] 472.A.III/IV.C.151.b (about 420 a.C.)

        Many authors have denied the existence of the DF in Greek, like
Sommer and Kuehner-Gerth. Those who have access to the book will find a
good resume of such position in Lasso de la Vega, JosŽ. "Sintaxis griega".
Madrid, 1968 pp. 554 ss. Such view is, in my opinion, not sound.
        There is a very interesting article by Hoekstra, A. "A note on the
dative of purpose (dativus finalis) in Greek." Mnemosyne 15 (1962): 15-23
that (together with some really dubious examples of *e)/cw* + Dat. from
Thucydides) discusses about 15 datives with such meaning in the poetry and
prose of Classical Greek, and concludes that, even if the interpretation of
some of them is dubious, there is no reason to doubt of the existence of
the construction in Greek (about the same thread, see Schenkenveld, D. M.
"Antiphon Or. V, 8: A case of dative of purpose?" Mnemosyne 1989 (1989):
474-476). To my mind there are much more examples of DF that the ones that
are usually described in the manuals. I have been collecting several
examples in scraps of papers or as annotations in many books I don't have
here at hand, but I remember 2 passages deserving at least some
consideration not quoted in the above mentioned works (I am sorry that I
couldn't check Cooper's Syntax who almost always have something interesting
to say):

        h)=san de\ eu)stalei=s te th=| o(pli/sei Th.3.22.2 (most probably
not to be interpretated as "in the o.")
        telei/wsis toi=s lelalhme/nois au)th=| para\ kuri/ou Eu.Luc.1.45

        I checked 8 different Spanish translations of Gal.5.1 and all of
them give a final translation like "para la libertad nos ha liberado/
libertado Cristo" except one giving a local interpretation.

Daniel Ria–o Rufilanchas
Madrid, Espa–a

B-Greek home page:
You are currently subscribed to b-greek as: []
To unsubscribe, forward this message to
To subscribe, send a message to

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:40:29 EDT