the case of the subject of an infinitival clause -- repost

Carl W. Conrad cwconrad at
Sun Sep 30 09:02:12 EDT 2001

I appreciate Chet's effort taken to send a typographically-corrected copy
of the message. To this nice little piece of careful research I'd like only
to offer a couple footnotes.

At 7:09 AM -0400 9/30/01, Chet Creider wrote:
>[the original posting contained 6 typos in which the letter "V" was
>mistakenly typed instead of "N" -- a cognitive confusion I have which
>is perhaps caused by being left-handed; I've had the problem as long as
>I've been typing Greek and can't seem to catch it except by proof-reading
>(obviously not done well in this case)-chet]
>The following was touched off by the recent discussion involving AUTOS
>and attributive and so-called PREDICATE position, but involves a different
>matter, the case of the subject of infinitives.  AUTOS, however, plays
>a role in some of the examples.
>I apologise for the length of the question, but since at least one
>person on the list makes use of North & Hillard, I thought it best
>to precede my question with a discussion of an error they make.  Readers
>wishing to skip to the question itself may search for "II." and should
>ignore section III.
>North&Hillard (p.50):
>[24. II Verbs of saying and thinking may also be followed by the
>Accusative with the Infinitive, as in Latin.
>	But in Greek if the subject of the Infinitive is the same
>as the subject of the principal verb, it is expressed only where
>emphasis or contrast is required, and, if expressed, it is in the
>The last clause is not true (see below for examples).  N&H don't give
>any examples, but with a little work, I think I have tracked down the
>source of the error:

I assume that the "North and Hillard" you're citing is the veteran standard
_Greek Prose Composition_ which was my own textbook in Greek Prose Comp
back in 1959-60. It's not a book I'd want to use as a reference grammar in
any case, but certainly the principle is mis-stated. What DOES need to be
clearly stated is that the subject of an infinitive, if identical with that
of the introductory clause, is usually unexpressed and nominative, and in
that case its modifiers, adjectives or predicate words, are also in the
nominative. That is precisely the case, I think with the disputed sentence
here: OUK EFH AUTOS ALLA NIKIAN STRATHGEIN, where AUTOS is indeed not the
subject itself but an intensive pronoun (I wouldn't call it an adjective,
but perhaps the Greek term for it would be EPIQETON) modifying an implicit
nominative subject. And yes, I think that the principle in this case is
that the nominative of the clause in the direct speech is retained.

>[from the syntax part of Abbott & Mansfield, _A Primer of Greek Grammar_
>(p.207): "The Subject of the Infinitive, if it is the same as that of the
>principal Verb, is placed in the Nominative:
>	OUK EFH AUTOS ALLA NIKIAN STRATHGEIN He said that he was not in
>	command, but Nicias.]
>My copies of these works are Duckworth reprints and no original date of
>publication is given, but in Goodwin's 1894 (2nd edition of first edition
>of 1879) _Greek Grammar_ the original of the example given by A&M is
>given.  Since Goodwin & Gulick (1930) does not change this passage, I'll
>refer you to the page/paragraph of this version (G&G), pp.202-3, para.
>923: "When the subject of an infinitive is omitted because it is the same
>as the subject nominative of the leading verb, adjective words and nouns
>which would agree with the omitted subject are assimilated to the preceding
>nominative, E.g....OUK EFH AUTOS ALL' EKEINON STRATHGEIN he (Cleon) said
>that not (he) himself, but he (Nicias) was general...AUTOS being adjective
>and EKEINOS substantive (Thuc 4,28)."
>[NB: Goodwin is _not_ saying that autos is a subject; rather he is
>using this as an example of nominative agreement with an unexpressed
>subject of an infinitive.]
>It seems clear that the British school grammar error arose from a
>misanalysis of this sentence.  How or when this was first done, I have
>no idea.
>Smyth discusses the same sentence on p.439, para. 1974a (of the 1958
>revised edition) and explicitly states that " not the expressed
>subject of the infinitive, but AUTOS of direct discourse (AUTOS TUPHSEIS,
>In sum, the rules governing agreement when the subject of the infinitive
>is the same as a main clause subject seem to be these:
>(1) if the subject of the infinitive is not expressed, agreeing elements
>(predicate nouns, adjectives, participles, etc.) are in the nominative.
>(2) the only exception to this is when the subject of the infinitive is
>emphasized, and in that case, it is in the accusative and any agreeing
>elements are also then in the accusative.  Smyth states this very clearly
>(p. 439, para 1974): "A pronoun subject of the infinitive, if...identical
>with the subjectof the main verb, is generally expressed when emphatic,
>and stands in the accusative (cases of the nominative are rare and

Since Smyth #1974 has been cited, I think it only fair to note that the
ellipsis at the end of the citation should be completed with a notable
exception: " ... ; but the indirect reflexive SFEIS stands in the
nominative or accusative." And he illustrates that nom. SFEIS with Dem
(the speaker and the other Lacedemonians) BUT THEY (the Toroneans) RATHER

>In addition to the examples Smyth gives in that paragraph (plus some
>Homeric examples I give below), there is a lovely example from Plato on the
>following page in Smyth (p.440, para. 1976): "HMIN DE POIOUSI DOKEIN SFAS
>PANTODAPOUS FAINESTHAI they manage it so that they seem to us to appear
>in various forms (P.R. 381 e)"
>Here the subject of the main clause (they) is not expressed but is the
>same as SFAS, which is in the _accusative_, and the agreeing element,
>PANTODAPOUS, is of course also accusative.
>The above rules appear to have been constant at least from Homer through the
>Attic period.  However, there is at least one exception in the New Testament:
>HUCOMHN ANAQEMA EINAI AUTOS EGW "I wished myself to be (the) accursed (one)"
>(Rom. 9.3)
>Does anyone know of any others?

Just a couple quibbles here (and that's all they are--quibbles): (1) I
don't think I'd speak so much of " normative rules" here so much as of
"prevalent practice"; (2) The sentence in Rom 9:3 could as easily (or
perhaps even better) be understood with AUTOS EGW as the actual subject of
HUCOMHN rather than of the infinitive EINAI: "I myself would have preferred
to be accursed"--at any rate, I I don't think that I myself would quite
call it an "exception" to a "rule."

>Here are some counterexamples to North & Hillard's claim from Homer:
>	but of those others it is I who is better by far
>	since I think I am not so very inexperienced, born and raised
>	in Salamis
>	for I say neither am I forgetful of valour

Another little quibble here: although N&H do mis-state the principle, I
don't think they intended to describe the usage of Greek as a whole but
only to describe Attic prose practice for students wishing to write Attic

Carl W. Conrad
Department of Classics, Washington University (Emeritus)
Most months: 1647 Grindstaff Road/Burnsville, NC 28714/(828) 675-4243
cwconrad at OR cwconrad at

More information about the B-Greek mailing list