On the accepted general meaning of μή

[NateD26]

In university and in most grammar books, it's being staunchly taught that the employment
of μή and not οὐ in relative clauses and substantive participles like ὁ μὴ ποιῶν, ἃ μὴ οἶδα,
etc., must have some conditional or subjective meaning to it and if it were objective
or matter of fact, it would have οὐ. I wonder whether the case is always thus.

There's this book I've linked to in an earlier thread in which the author argued, albeit vehemently
and with a bitter, anti-German tone, that μή had been employed just as often as οὐ without any
conditional, abstract, or subjective meaning hidden in it. I've not finished the book yet
but from what I could gather, his theory is that all instances of μή stem from the original prohibitive
meaning; that by employing μὴ -- a stronger negative -- and not οὐ, the intended meaning of the author
is to present an action or result "as bad or to be avoided".

I'll kick off this thread by giving this sentence as the first example from Plato's Apology
where my teachers and grammar books are adamant in their conviction that there's a conditional
meaning in the relative clause:

ἔοικα γοῦν τούτου* γε σμικρῷ τινι αὐτῷ τούτῳ σοφώτερος εἶναι,
ὅτι ἃ μὴ οἶδα οὐδὲ οἴομαι εἰδέναι. (21d)

* τις τῶν πολιτικῶν (21c)

[my clunky translation]
It seems to me then, that at least in this very small thing I am wiser than this politician,
namely, that what I do not know, I do not even pretend to know.

Must it have some hidden conditional meaning as in εἰ μὴ οἶδα...? That would seriously
reduce the strength of Socrates' statement, who felt very strongly about people pretending
to hold some knowledge they do not actually possess.

[pster]

Sidgwick's distinction between negative statements and negative conceptions seemed reasonable and brave if not perfect. I think it is natural to include the abstract on the conception side. And I don't think of the abstract as weaker. Indeed, perhaps quixotically, I prefer to think of it as stronger. So I don't see any problem as far as this abstract Plato example is concerned. Maybe you can give English translations for both readings? Interesting thread though. I have some good essays on early Plato, so I'll see if there are any discussions of that passage. There may very well be. But it is late and I will abstain from drawing further conclusions for the time being.

[NateD26]

Sidgwick's distinction between negative statements and negative conceptions seemed reasonable and brave if not perfect. I think it is natural to include the abstract on the conception side. And I don't think of the abstract as weaker. Indeed, perhaps quixotically, I prefer to think of it as stronger. So I don't see any problem as far as this abstract Plato example is concerned. Maybe you can give English translations for both readings? Interesting thread though. I have some good essays on early Plato, so I'll see if there are any discussions of that passage. There may very well be. But it is late and I will abstain from drawing further conclusions for the time being.

Can you please post the reference to Sidgwick's treatment of the negatives?

In class (and in commentaries), the accepted translation of the Plato's Apology quote is
with a condition fixed in the relative clause:
...namely, that what I do not know, if I do not know it, I do not even pretend to know.

To me, this condition limits the force of his statement. This is just one sentence off the top
of my mind. I'm sure there are examples to support each reading, and perhaps there's
the bias of the author, making each example he cited fit to his own theory. I'm still
not convinced that in my particular example there should be any abstraction or condition
hinted by it.

[C. S. Bartholomew]

I

I'll kick off this thread by giving this sentence as the first example from Plato's Apology
where my teachers and grammar books are adamant in their conviction that there's a conditional
meaning in the relative clause:

ἔοικα γοῦν τούτου* γε σμικρῷ τινι αὐτῷ τούτῳ σοφώτερος εἶναι,
ὅτι ἃ μὴ οἶδα οὐδὲ οἴομαι εἰδέναι. (21d)

* τις τῶν πολιτικῶν (21c)

[my clunky translation]
It seems to me then, that at least in this very small thing I am wiser than this politician,
namely, that what I do not know, I do not even pretend to know.

Must it have some hidden conditional meaning as in εἰ μὴ οἶδα...? That would seriously
reduce the strength of Socrates' statement, who felt very strongly about people pretending
to hold some knowledge they do not actually possess.

Nate,

Perhaps the problem is over interpretation of the meta language: objective/subjective. The first clause ὅτι ἃ μὴ οἶδα specifies a set of knowledge which is indeterminate, "what ever I don't know." When some token of knowledge comes a long it is tested against the question "do I know it?" if the answer is NO then the the second clause is true οὐδὲ οἴομαι εἰδέναι. You certainly understand this but I am just laying the ground work. It is the indeterminate nature of this set that makes μὴ the preferred particle. There isn't anything "hidden" about the conditional nature of ὅτι ἃ μὴ οἶδα. We are not reading something into this to say it is conditional.

Cooper takes the same approach using the same meta language is the old Germans, LSJ, etc. However, he does demonstrated that οὐ can be used in a similar context. His explanation of this is difficult to summarize since I don't totally understand it. I will post later if I can make sense out it.


C. Stirling Bartholomew

[C. S. Bartholomew]

Cooper actually addresses this text (v. 2, #67.4.2.A pp. 1099-1100).

When relative sentences have a generic, and so sometimes plainly hypothetical or conditional sense, there negative is μὴ.

Pl.Ap 21d ἃ μὴ οἶδα (i.e. εἴ τινα μὴ οἶδα)

C. Stirling Bartholomew

[NateD26]

Thanks, CSB, for your explanation and the Cooper reference. Smyth deals with this sentence as
well in 2507 and also sees a condition in it, making a distinction between definite (negative οὐ)
and indefinite antecedent (negative μή), which is the distinction we've been taught in class and
precisely the point the author of the book I've linked to in my OP tried to demonstrate how
unsubstantiated it can be in actual writing, though his translations of the cited examples are
probably fashioned to suit his narrative, with simplistic and repetitive arguments.

[C. S. Bartholomew]

Thanks, CSB, for your explanation and the Cooper reference. Smyth deals with this sentence as
well in 2507 and also sees a condition in it, making a distinction between definite (negative οὐ)
and indefinite antecedent (negative μή), which is the distinction we've been taught in class and
precisely the point the author of the book I've linked to in my OP tried to demonstrate how
unsubstantiated it can be in actual writing, though his translations of the cited examples are
probably fashioned to suit his narrative, with simplistic and repetitive arguments.

Nate,

On page 15 the author cites Oedipus Coloneus ...

S.OC Line 73 Καὶ τίς πρὸς ἀνδρὸς μὴ βλέποντος ἄρκεσις;

... and claims that ἀνδρὸς μὴ βλέποντος has a definite referent. But does it? Obviously it is at some level a reference to Oedipus but the form it takes is indefinite and the question on the surface is a general one about blind men. I think the author is confounding surface structure with meaning and inference. We automatically infer that Oedipus is the blind man. But the surface structure is still indeterminate, any member of a class of men who are blind.


C. Stirling Bartholomew

[pster]

Can you please post the reference to Sidgwick's treatment of the negatives?

http://archive.org/stream/cu31924021601 ... 5/mode/2up

[NateD26]

Can you please post the reference to Sidgwick's treatment of the negatives?

http://archive.org/stream/cu31924021601 ... 5/mode/2up

Thanks, pstr. That's a neat, concise summary of Smyth's sections.

Nate,

On page 15 the author cites Oedipus Coloneus ...

S.OC Line 73 Καὶ τίς πρὸς ἀνδρὸς μὴ βλέποντος ἄρκεσις;

... and claims that ἀνδρὸς μὴ βλέποντος has a definite referent. But does it? Obviously it is at some level a reference to Oedipus but the form it takes is indefinite and the question on the surface is a general one about blind men. I think the author is confounding surface structure with meaning and inference. We automatically infer that Oedipus is the blind man. But the surface structure is still indeterminate, any member of a class of men who are blind.


C. Stirling Bartholomew

That's exactly the accepted treatment of μὴ with the participle. Here, it does seem
there's a connection we make in our minds to Oedipus as the blind man, but the question
is indeed general. Sir Richard Jebb sees a condition in μὴ βλέποντος, which is usually
the way we were instructed in class to read a participle with μὴ:

μὴ βλέποντος, not οὐ, since the blindness is a condition: "if he has not sight."

My question would be if it is always the case, and if some of the examples in the book
I referenced do present a definite, matter-of-fact statements, rather than conceptions.

In a previous thread, you've mentioned line 540 from Sophocles' Ajax and I'm not sure
what was your final reading of the quote, since you questioned whether it's an example
of an interrogative expecting negative reply. Sidwick (p.71) wrote this quote slightly differently
and read it as a virtual negation of the negative verb, that is, a question expecting negative
reply.

[pster]

In class (and in commentaries), the accepted translation of the Plato's Apology quote is
with a condition fixed in the relative clause:
...namely, that what I do not know, if I do not know it, I do not even pretend to know.

Nate, I disagree with this. The Loeb translation is probably by far the biggest seller: "...that what I do know know, I do not think I know either." And Smyth's translation is: "whatever I do know know, I do not even think I know." Neither of them has the extra bolded clause that you have inserted. So I don't know on what you are basing your claim about "the accepted translation".

I think you are adding an extra condition. The condition is the first part. It is not some extra thing that is lurking implicit and that turns two clauses into three when properly understood or translated. Indefinite clauses often get treated like protases of conditionals in grammars of different languages (not just Greek). So, "Whoever is a man is mortal" can be thought of as the conditional "If someone is a man, he is mortal." Often they follow the same sequence of tenses. So they get discussed together. In Greek, all the protases get mh, and similarly, indefinite clauses get mh. Ordinary subordinate clauses (as opposed to indefinite ones), like the one in the first part of "The man who is wearing the green shirt is mortal" don't work the same way because it is not easy or even possible to associate conditionals with them.

[NateD26]

In class (and in commentaries), the accepted translation of the Plato's Apology quote is
with a condition fixed in the relative clause:
...namely, that what I do not know, if I do not know it, I do not even pretend to know.

Nate, I disagree with this. The Loeb translation is probably by far the biggest seller: "...that what I do know know, I do not think I know either." And Smyth's translation is: "whatever I do know know, I do not even think I know." Neither of them has the extra bolded clause that you have inserted. So I don't know on what you are basing your claim about "the accepted translation".

I think you are adding an extra condition. The condition is the first part. It is not some extra thing that is lurking implicit and that turns two clauses into three when properly understood or translated. Indefinite clauses often get treated like protases of conditionals in grammars of different languages (not just Greek). So, "Whoever is a man is mortal" can be thought of as the conditional "If someone is a man, he is mortal." Often they will follow the same sequence of tenses. So they get discussed together. In Greek, all the protases get mh, and similarly, indefinite clauses get mh.

I should have put it in parenthesis, as Smyth did. Apologies. It's just another reading. A condition
or general/indeterminate/indefinite as opposed to specific/particular/definite statement.
Why should the inclusion of μή instead of οὐ immediately turn the relative clause, substantive
participle, etc. indefinite? Do we have enough evidence in the extant corpus to support
such reading, or just as much evidence to dispute it? I'm not disagreeing completely with this
accepted reading, just want to figure out whether other kind of readings, here or elsewhere, are possible.

[pster]

You have raised a bunch of questions that go in different directions. Socrates' famous claim has been the subject of many PhD theses. The book that you linked to at the outset looks old, tendentious, and frankly not worth wasting one's time on.

You ask:
"Why should the inclusion of μή instead of οὐ immediately turn the relative clause, substantive
participle, etc. indefinite? Do we have enough evidence in the extant corpus to support
such reading, or just as much evidence to dispute it?"

But these are two different questions. I'm not sure what kind of answer you would even want for the first. Grammatical rules are abstracted from practice. As for the second, let's stick with relative clauses for the moment. The mh doesn't make it indefinite. If you can easily turn it into a conditional, then it is probably indefinite. And so mh would be used. If you can't turn it into a conditional, then it is probably not indefinite. And so ou would be used. I think somehow you are over-imagining things when you speak of "hidden conditional meaning". In general, I don't think there is anything hidden at all in these cases. The only thing that is making things seem obscure is the choice of Socrates' famous enigmatic claim as an example. "Indefinite" just means general, ie applying to a class of things.

Libraries are filled with plenty of old books that provide unorthodox accounts of things. If you can tell us why this particular account is worth thinking about, that would help. Did mh and ou bother you before you read this book?

[C. S. Bartholomew]

Nate,

There are exceptions of another kind, cases where οὐ is used where we might expect μὴ.

Plat. Apol. 25b

[25β] ... οὐχ οὕτως ἔχει, ὦ Μέλητε, καὶ περὶ ἵππων καὶ τῶν ἄλλων ἁπάντων ζῴων; πάντως δήπου, ἐάντε σὺ καὶ Ἄνυτος οὐ φῆτε ἐάντε φῆτε: πολλὴ γὰρ ἄν τις εὐδαιμονία εἴη περὶ τοὺς νέους εἰ εἷς μὲν μόνος αὐτοὺς διαφθείρει, οἱ δ᾽ ἄλλοι

[25b] ... Is it not so, Meletus, both in the case of horses and in that of all other animals? Certainly it is, whether you and Anytus deny it or agree; for it would be a great state of blessedness in the case of the youth if one alone corrupts them, and the others do them good. But,

Cooper (v. 1, 67:4.1.B p1099) cites Plat. Apol. 25b ἐάντε σὺ καὶ Ἄνυτος οὐ φῆτε ἐάντε φῆτε: as an example of οὐ being used where we might expect μὴ.

C. Stirling Bartholomew

[NateD26]

Libraries are filled with plenty of old books that provide unorthodox accounts of things. If you can tell us why this particular account is worth thinking about, that would help. Did mh and ou bother you before you read this book?

I didn't have any problem because the distinction explained in Smyth & Sidwick was fixed in my mind,
but upon reading this author's book (found randomly on archive.org with no intention of searching
for a counter-theory), I've started asking myself these questions, and I agree his view is on the
fringe of Classical studies, though he nonetheless makes convincing arguments, if a bit repetitive.

Nate,

There are exceptions of another kind, cases where οὐ is used where we might expect μὴ.

Plat. Apol. 25b

[25β] ... οὐχ οὕτως ἔχει, ὦ Μέλητε, καὶ περὶ ἵππων καὶ τῶν ἄλλων ἁπάντων ζῴων; πάντως δήπου, ἐάντε σὺ καὶ Ἄνυτος οὐ φῆτε ἐάντε φῆτε: πολλὴ γὰρ ἄν τις εὐδαιμονία εἴη περὶ τοὺς νέους εἰ εἷς μὲν μόνος αὐτοὺς διαφθείρει, οἱ δ᾽ ἄλλοι

[25b] ... Is it not so, Meletus, both in the case of horses and in that of all other animals? Certainly it is, whether you and Anytus deny it or agree; for it would be a great state of blessedness in the case of the youth if one alone corrupts them, and the others do them good. But,

Cooper (v. 1, 67:4.1.B p1099) cites Plat. Apol. 25b ἐάντε σὺ καὶ Ἄνυτος οὐ φῆτε ἐάντε φῆτε: as an example of οὐ being used where we might expect μὴ.

C. Stirling Bartholomew

This is an interesting peculiarity, Stirling. What does Cooper suggest as an explanation?
I'd assume οὐ φῆτε is treated as a single constituent, to deny, rather than a negation of a verb,
to not say, in which case there's no need for an explanation.

[C. S. Bartholomew]

This is an interesting peculiarity, Stirling. What does Cooper suggest as an explanation?
I'd assume οὐ φῆτε is treated as a single constituent, to deny, rather than a negation of a verb,
to not say, in which case there's no need for an explanation.

Nate,

Cooper didn't comment on each text, it was buried in a whole list of cases where οὐ was used in a conditional statement. I'm not sure how to answer your question if οὐ φῆτε is treated as a single constituent. οὐ negates the verb in an either or negative/positive condition.

CSB

[NateD26]

This is an interesting peculiarity, Stirling. What does Cooper suggest as an explanation?
I'd assume οὐ φῆτε is treated as a single constituent, to deny, rather than a negation of a verb,
to not say, in which case there's no need for an explanation.

Nate,

Cooper didn't comment on each text, it was buried in a whole list of cases where οὐ was used in a conditional statement. I'm not sure how to answer your question if οὐ φῆτε is treated as a single constituent. οὐ negates the verb in an either or negative/positive condition.

CSB

Well, it does negate the verb but it seemed at first to take its own meaning before being placed
in the condition. This commentary seems to read it in a similar manner.

I also couldn't find initially instances where we would have
μή with φημί in conditions, but LSJ did cite a couple.

[pster]

What is Cooper? Is there a link?

[C. S. Bartholomew]

What is Cooper? Is there a link?

Attic Greek Prose Syntax:
K. Kruger, Guy L. Cooper
University of Michigan Press

Very little of vol one available for preview Google.

CSB