Can anyone think of any pairs of words in Greek that are distinguished purely by a rough vs smooth breathing?
Two have occurred to me already: οἷος & οἶος ; ὅρος & ὄρος. I’d be grateful if anyone can come up with any more.
(Note that something like εἰς vs εἷς does not quite count because the words differ not only in aspiration but also in accentuation.)
There’s also the issue of whether they are the same, but the vowel quantity may differ.
Off the top of my head a pretty common one, of course due to crasis, is the Attic
ἅνθρωπος [ie, ὁ + ἄνθρωπος] vs ἄνθρωπος
ἔργμα ἕργμα
ὥριος ὤριος
ἔω ἕω
αὐτοῦ αὑτοῦ
εὖ εὗ
εἴλησις εἵλησις
ἄν ἅν
ὥσπερ ὤσπερ
εἶ εἷ
ἁλιάς ἁἀλιάς
αὐτός αὑτός
ἀλεία ἁλεία
ἀλίη ἁλίη
ὁμόω ὀμόω
ἤμων ἥμων
ἔμεν ἕμεν
ἕσσα ἔσσα
ἔξω ἕξω
ἰός ἱός
ἅλιμος ἄλιμος
οἶος οἷος
ὁροφύλαξ ὀροφύλαξ
ἑλλός ἐλλός
οὐκ οὑκ
ἐναχῶς ἑναχῶς
ἵσμα ἴσμα
ἅζω ἄζω
ἅλινος ἄλινος
ἐξάγιον ἑξάγιον
ὄρος ὅρος
ὄσσα ὅσσα
ὅλονθος ὄλονθος
εἴλη εἵλη
ὦ ὧ
ἅλσις ἄλσις
ἔπω ἕπω
εἰλέω εἱλέω
ὠρεῖον ὡρεῖον
ἵ ἴ
ἐπωπίς ἑπωπίς
ἐδανός ἑδανός
ὤρα ὥρα
ἦς ἧς
ὀρικός ὁρικός
ἤκιστος ἥκιστος
ἀλίτης ἁλίτης
ὦρος ὧρος
ἅρμα ἄρμα
ἐξέμεν ἑξέμεν
ἑκάς ἐκάς
ἴεσις ἵεσις
ἧλος ἦλος
οὖ οὗ
ἦκα ἧκα
ἅλμα ἄλμα
ἅλυσις ἄλυσις
ἑή ἐή
ὄρριον ὅρριον
ἐξάριθμος ἑξάριθμος
ἐνίζω ἑνίζω
ἑκτός ἐκτός
ἑξαπλόω ἐξαπλόω
ἀλεάζω ἁλεάζω
αὔω αὕω
ἔνος ἕνος
ἐνί ἑνί
ἐσία ἑσία
ὁ ὀ
ὀδός ὁδός
On crasis: of course this is its own ball-game.
On differences in vowel quantity: would be a bit surprised if it factors often, but I hadn’t thought of it so thanks! Will keep a watch out.
Wow!
Mind explaining how you prepared this? If, as I’m inclined to guess, you ran some code on Perseus’ Morpheus, I’d love to have a look at it.
As for the list itself: there are of course plenty of entries I’m not too sure about. e.g. ὤσπερ: I could only find this used in Herodas, where it is simply psilotic ὥσπερ. Others are not securely attested. And is there any dialectical admixture where we could expect ἄν (as opposed to Doric κα) and ἅν (Doric rel. pronoun), to both be usable together?
Yet for all that, this answer is much much better than anything I expected or hoped for. It gives me plenty of certain pairs to add to our list: ἦκα & ἧκα, ὦ & ὧ, ἔξω & ἕξω, ὀδός & ὁδός, ἐνίζω & ἑνίζω, εἴλησις & εἵλησις, ἔργμα & ἕργμα, ὦρος & ὧρος, and I haven’t explored the list item-by-item yet (just checked up on the ones that struck a chord right away). It has the added virtue, I think, of being comprehensive (for 5th century Attic at least). So thank you!
I ran some code against a digitized LSJ from the internet archive. It didn’t take much, as I already had libraries prepared. It is not comprehensive, however, as I only considered dictionary headwords. Morphological forms no doubt contain other collisions.
results = []
ug = unigreek.UniGreek()
for kk, vv in lsj_headwords.items():
if len(vv['entries']) > 1:
entry_raw_heads = []
for ee in vv['entries']:
entry_raw_heads.append(lsj_raw_entry(ee)['headword'])
found_accent = False
smooth_hw = None
rough_hw = None
for hw in entry_raw_heads:
if hw.count(')') > 0:
found_accent = True
smooth_hw = hw
rough_hw = hw.replace(')', '(')
break
for hw in entry_raw_heads:
if hw == rough_hw:
entry_greek_heads = []
for hw in entry_raw_heads:
entry_greek_heads.append(ug.to_greek_with_accents(hw.lower()))
results.append(entry_greek_heads)
break
ἄν and ἅν are both used in “Doric” lyric (incl. tragedy). But ἄν is short, ἅν long.
With many of the others the difference is merely orthographic or dialectal. And in context there’d be no confusion with any of them. I think this is a silly game.
And so, from hour to hour, we ripe and ripe,
And then, from hour to hour, we rot and rot;
And thereby hangs a tale.’
Ben Crystal reciting the above.
I think it’s linguistically interesting that this sort of pun can’t be done in Greek. The similar h-dropping lists for English seem to show more room for confusion. [1] [2]
εναχως might be good for a pun (one-way/nine-ways), but it’s the only one I see on the list.
I’m not quite sure I agree with this.
What do you think for instance of the work done (a long time ago) to create accent minimal pairs (words distinguished by accent alone), such as the lists Stephanus, Goettling, Jelf produced?
In particular, the rarity of some of the words involved here may have resulted in copyist scribes regularizing the breathing (unaware of the meaning of the rarer word, or that such a word existed), leaving us with a text that is at best unintelligible and at worst carries a wrong but credible meaning. As scholars it would be necessary for us to recognise the minimal pair in question - which involves knowing the rare word as well as the common - to correct the text. (And moreover, knowing the rare word well enough that it come to mind, unprompted, at such a juncture: which is rather harder than simply recognising and understanding it when you read it.)
Then again, for intermediate students, or in general those readers who do not yet know all those words (even passively), there is the further benefit of ensuring that when they run into them they will not misinterpret and carry on (through failure of distinguishing the breathing properly) (but will instead stop and look up the word, as they’d have to for any other word they don’t know). This - the prevention of misinterpretation by intermediate students - seems to me the likeliest reason why the lists distinguishing accents may come in handy: between e.g. βασιλεία and βασίλεια, the greatest danger is that a student may simply not recognise the misplaced accent as they read: I suggest the same may be true (to an extent) for a breathing. They are not always better emphasised by teachers than accents (see e.g. Italians on the whole).
(Yes, this is a “game” rather than a philologically useful (let alone necessary) tool. It certainly didn’t actually arise out of my desire to be able to recognise wrong breathings when I come across them in manuscripts. Still, I invite a response!)
To answer your question: I have no interest at all in lists of words allegedly distinguished by accent alone. Such words are all distinguished by something else as well, something more important (meaning, for one). Inexperienced readers do sometimes confuse βασίλεια and βασιλεία. But that provides an opportunity for them to learn some elementary morphology, and to distinguish short vowels from long ones. Always best done in a concrete setting. Sometimes a student will mistake ἄν (long α, = ἐάν) for the modal particle, but that again is a learning opportunity, registering the positional difference along with the functional.
Nowadays such decontextualized lists can be generated in a mattter of minutes, as Joel demonstrated with breathings, but I can’t see this as a real benefit. Students would learn far more by reading texts and using a dictionary.
As to copyists hypothetically getting breathings wrong, that’s a non-issue. Most manuscripts are full of corruption, and editors learn not to rely on breathings, which are only a secondary accession to the text in any case, generally absent from ancient manuscripts.
Sure but what does that have to do with the argument here? Of course editors “learn not to rely on breathings” in the course of learning textual criticism; indeed I am sure any reader of your post who did not know that before now knows that; yet that awareness is not the key discriminator whether he/she will e.g. be able to recognise the need to emend a certain οἷος to οἶος.
(Note that I chose οἷος - οἶος because that is an example I have encountered a couple of times in my own training. I accept that the commonality of οἶος vitiates the example - one could certainly say that until a student knows vocabulary like οἶος well, simply through general learning, he should not concern himself with textual matters. But I doubt this is the case for some other examples on the list.)
Your holistic approach is commendable but I’d ask how far you extend it. It is all well and good to say “reading first!”, but would you prefer “reading first, syntax after” or “reading first, syntax never”? After all - just as one absorbs lexical issues like these minimal pairs from reading experience, to an extent - they also absorb grammar and syntax; does that mean that you counsel them to avoid ever recensing syntax, instead commanding them to just keep on reading? If so, why not apply that to morphology, which on some level you could also claim a reader absorbs just by reading? I noticed you advocate the use of dictionaries, but why - surely the semantic range of a word is best learned on the trot?
If your approach does not extend itself to forbidding syntax study, then I suppose we come to an agreement. That is, the reason I myself don’t think the above list is of actual pedagogical use is because there are other things students could be learning with their time in preference - the kind of revision it provides is too niche to be regularly useful. That doesn’t mean it has no value but simply that it is not the tool of preference that we would actually spend time on.
Extensive reading and interaction with the language nails down both morphology and syntax. That doesn’t mean there is no value in memorizing paradigms and vocabulary, contra the more radical Krashenites, but simply that such must be done in the framework of using the language. Then “difficult” forms and similar words tend to be learned in context, the result being that the student rarely continues to have difficulty.
For pedagogical purposes, this is almost useless. The very few useful words in my above list are in the Morwood ψευδεῖς φίλοι list, found in his grammar.
But again, I find it linguistically interesting there there can be a lot of confusion caused by h-dropping in English, between very basic and common words, but there appears to be almost none in Greek, as almost all of the words in my list are very rare.
Chommoda dicebat, si quando commoda vellet
dicere, et insidias Arrius hinsidias.
et tum mirifice sperabat se esse locutum,
cum quantum poterat dixerat hinsidias.
Credo, sic mater, sic liber avunculus eius,
sic maternus avus dixerat atque avia.
Hoc misso in Syriam requierant omnibus aures:
audibant eadem haec leniter et leviter,
nec sibi postilla metuebant talia verba,
cum subito affertur nuntius horribilis
Ionios fluctus, postquam illuc Arrius isset,
iam non Ionios esse sed Hionios.
No h-dropping there!
I don’t think I’ve ever come across genuine confusion in English, just some contrived puns. It’s just a dialectal feature like any other, only easily represented in writing.
I had a look at that list earlier today. Seems to be missing ἔξω & ἕξω which is probably the most important minimal pair (for a beginner to learn) of this kind. Also maybe a few others (ἧκα & ἦκα?) but I don’t remember because I don’t have the book with me.
Since what I hoped for from the thread has been accomplished, I’ll leave things here. Perhaps for a discussion on certain (less directly relevant) issues that have come up I will open a thread sometime in the Academy forum.