Bers, Victor. 2009. Genos Dikanikon: Amateur and Professional Speech in the Courtrooms of Classical Athens. Hellenic Studies Series 33. Washington, DC: Center for Hellenic Studies. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Bers.Genos_Dikanikon.2009.
Chapter 7: Tactics, Amateur and Professional
An exceptionally diffident amateur litigant might have made the mistake of hedging with που (even one occurrence might have been too many), but it seems likely that in this detail at least most idiôtai and professionals would have followed the same practice.
Two aspects of ἀτάρ are noteworthy. Its semantic force, particularly in Attic usage, is to mark “a break-off, a sudden change of topic” (52), and it is a particle given extra prominence by its invariable position as the initial word in a sentence or clause. I suggest that the professionals avoided the word because it risked dramatizing the speaker’s astonishment or indecision in the midst of his utterance, perhaps brought on by a failure to prepare his speech adequately. ἀλλὰ μήν is common in the attested orators,  who (I think) heard the usage as allowing for a switch of topic, even to signal a presumably spontaneous idea,  while still marshalling a more dignified parade of assertions.
Presumably logographoi were careful not to use ἦ μήν in a way that suggested sloppiness in a legal matter. Amateur speakers were unlikely to be so fastidious, and their usual linguistic habits  would put them at risk of seeming too casual with oaths.
Someone who was not a “budding speaker” might or might not have picked up this verbal tic. We have, then, a usage that an idiôtês might miss, and thereby demonstrate his lack of training; or he might try to imitate this feature in too obvious a manner, say by wearing a smug expression and looking around for his supporters to signal their admiration. Denniston’s way of putting it suggests that the sequel to that opening particle was not of much importance, but the stylistic choices made by logographoi suggest they strove in particular to avoid uses with so strong a smell of the emotional, the diffident, or the colloquial as to mark the speaker as disrespectful of the state’s judicial process. These evitanda are, I am arguing, precisely the pitfalls awaiting an idiôtês coming before an Athenian jury without the armor of a professionally written speech in his head or in his hands or, at a minimum, friends and relatives able to help him meet the challenge.
Oaths and Exclamations
It may be significant that Antiphon relaxed his normal prosecutorial practice in a case that by category involved more than a single individual, and that he makes his claim that many others in the city were also victims: the wholesale change seen in Demosthenes might be a symptom of the same phenomenon.
MacDowell’s translation uses quotations marks to signal the imaginary change of speaker: “But we must go on fighting until we’ve beaten the Spartans and their allies”; “But I don’t think we’re equipped for that; and if we do accomplish it, what do you think the barbarians will do to us in our turn when we’ve done it?” 
τυροῦ κάκιστον ἀρτίως ἐνήρυγεν
ὁ βδελυρὸς οὗτος.
By Zeus, he did it, plain as day. This filthy dog just let rip the most awful cheesy belch at me!
Bdelycleon remonstrates with his father (919–920):
πρὶν ἄν γ᾽ ἀκούσῃς ἀμφοτέρων.
By the gods, Dad, don’t convict him before you’ve heard both sides!
And just as the prosecuting dog sums up and takes his seat, Philocleon is quite beside himself (931–934):
ὅσας κατηγόρησε τὰς πανουργίας.
κλέπτον τὸ χρῆμα τἀνδρός. οὐ καὶ σοὶ δοκεῖ,
ὦ ᾽λεκτρυών; νὴ τὸν Δί᾽ ἐπιμύει γέ τοι.
Wow! What a pile of crimes he’s denounced! Here’s a lump of larceny don’t you think, Rooster? By Zeus, the rooster’s winking “Yes”!
The prosecutor knows to avoid these exclamations; the defendant can, in any event, only say “Bow!” and hope the parade of his pups will mollify the one-man jury played by Philocleon. Real speakers – impatient, indignant, frightened – probably were not always so restrained. Even logographoi might break out with an utterance, not even a word in the standard sense, suggesting triumph or despair, but only Demosthenes dared leave one in a published text, and even he identified the cry as not precisely his alone. Once it is a sort of moan (in translation I use the inadequate “Oh!”)  his enemies’ politics inspire in brave men (23.210):
In Isaeus things begin to change: there are 15 instances according to Kühnlein, 14 by Dover’s count. Dionysius of Halicarnassus remarks that Isaeus was a transitional figure in the history of oratory,  and although Dionysius does not say so, I believe that the transition he effected was, in part, the introduction into the genos dikanikon of the more flamboyant style of the rhêtores who made frequent appearances in the Ecclesia. We then have comparatively large numbers for Aeschines (39 according to Kühnlein) and some 300 in Demosthenes. But even then, most of these oaths are not interjections that manifest the speaker’s emotion, but rather occur within the introductions that frame stretches of hypothetical speech attributed to persons other than the speaker (see Dover 1997:62–63). And there are only a few instances of oaths starting μά or νή in the latest orators of the Canon (Dinarchus, Hyperides, and Lycurgus).
λέγε θαρρήσας· ὡς τὰς σπονδὰς οὐ μὴ πρότερον παραβῶμεν. 
Two battlefield exhortations in Thucydides involve vehement denials that there will be invasions. The first is direct speech (4.95.2), the second is a clause within oratio obliqua (5.69.2):
… οὐ μη ποτέ τις αὐτοῖς ἄλλος ἐς τὴν γῆν ἔλθῃ.
… no one else will ever launch an attack on their territory. 
Plato uses the construction for Socrates’ manifesto (Apology 29d4–5):
Most striking is an appearance of the construction in the generally bloodless speech of the Laws. At 942c5–d3 the Athenian stranger speaks of the necessity for group obedience in military organizations:
The construction persists into New Comedy, as at Menander Samia 428, where an impatient character complains οὐ μὴ δύῃ ποθ᾽ ἥλιος· (“The sun will never go down!”).
But this is imagined speech, words that the speaker’ opponent, whose estate is under dispute, might have said under other circumstances. Most significantly, the negative command is not what would have been said in court, but in a private conversation, hence it is not strictly part of forensic speech (Bers 1997:141). 
It is very much in the speaker’s interests to present the defendant as pitiable in a bad sense. Part of that strategy – or perhaps simple exploitation of the facts – involves mentioning that Nicostratus was lachrymose during their meeting (§7: κλάων) and bears disfiguring marks that he will refuse to display, even if the jury asks him to, presumably from fear that the sight will trigger the jurors’ instinctive contempt for a body made ugly by servile labor. We might say that Apollodorus wants the jury to imagine the fierce obstinacy with which the defendant would refuse the request he is, in effect, trying to incite.
Given the pattern of usage, which includes tragic language and even presocratic poetry (Parmenides: see Collard 2005:378), Gildersleeve cannot be correct in characterizing the construction as a “vulgarity”; and the moralizing peroration tells us more about his social and ethnic views than about Greek. Still, he does seem to have caught the flavor of οὐ μή accurately enough: it is an expression that suggests the speaker raises his voice, or even stamps his foot, and that suggests a level of affect the skilled forensic speaker would work to avoid.
The rarity of σφόδρα before a pause does not look like an accident. My guess is that σφόδρα in that position was fairly common in excited colloquial speech and became an evitandum in professional writing because its rhythm was associated with overwrought vehemence, something like a speaker tacking on a “No, really!” after each of his assertions because he feels very little confidence in his rhetoric.
It is not surprising that Aristotle’s examples are all taken from Old Comedy, for it is the only genre of Greek literature in which diminutives can be found with any frequency. In this stylistic feature the canonical Attic orators certainly did observe Aristotle’s strictures: there are a handful in those preserved forensic speeches with an obvious political agenda, but elsewhere the speechwriters’ moderation amounts very nearly to total abstinence.
Wankel 1976 remarks that this is the sole instance of the word ἀνθρώπιον securely attested within a speech.  In On the Crown Demosthenes uses the word graidion (“old hag”) at §260 and at§261 arkhidion for a contemptibly low public office once discharged by Aeschines. In Against the Sophists, a work normally categorized as a sort of pamphlet, Isocrates uses λογίδια to refer to the theories of competitors who have claimed to offer instruction useful for the composition of lawcourt speeches, but to my knowledge there are no diminutives in any of Isocrates’ few actual forensic compositions (13.20).