Nicomachus Introduction to Arithmetic

Easter Greetings to Greek fans who are also math fans!

Despite consulting Smythe and Goodwin on Greek numbers, I am at a loss to make sense of the system behind the variety Nicomachus uses when substantively referring to numbers.

  1. Most (75%) numbers are preceded by the masculine singular article, as appositives to the expressed or (more often) implied “arithmos”, This makes sense to me.

  2. But I often see them preceded by a feminine or neuter article in the singular. When I find a suitable referent such as feminine “ekthesis” or neuter “skhēma”, I am relieved but still not 100% sure why he does this. Just as often I can find no suitable antecedent and am truly confused.

  3. Quite often the numbers are preceded by neuter plural article, perhaps as abstract quantities when no antecedent exists in the clause. This makes sense to me, but less sense than 1). Why does he sometimes do one and not the other? Elegant variation?

  4. Finally Nicomachus at least once just assigns a gender to some numbers, as in chapter nine below. Note the series with unusually determined numbers: all of them masculine except for 22, which is mysteriously feminine!

θ. Ἀρτιοπέριττος δέ ἐστιν ἀριθμὸς ὁ τῷ γένει καὶ αὐτὸς ἄρτιος ὤν, ἀντιδιαστελλόμενος δὲ ἰδικῶς τῷ προφρασθέντι ἀρτιάκις ἀρτίῳ, ὁ τὴν μὲν εἰς δύο ἶσα διαίρεσιν ἐπιδεχόμενος κατὰ τὸ κοινὸν γένος, τῶν μέντοι μερῶν ἑκάτερον εὐθὺς εἰς δύο ἶσα ἄτμητον ἔχων, οἷον
ὁ ϛ, ὁ ι, ὁ ιδ, ὁ ιη, ἡ κβ, ὁ κϛ,
οἱ ὅμοιοι· μετὰ γὰρ τὸ διχασθῆναι ἕκαστον τούτων ἀδίχαστα εὐθὺς τὰ μέρη εὑρίσκεται.

I have run into none of this trouble in Euclid. Has anyone here had experience with numerical treatises in Greek and can clear me up a bit? I have kept this post as brief as I could, but can easily supply examples for points 1) - 3) if anyone would like.

Eukharistō

Mark

Just about all I know about Nicomachus is in the Wikipedia article on him, and I have no text. When mathematicians turn mystic, I turn off (making an exception for Pythagoras himself). But maybe your answer lies there?

In the excerpt you quote, however, I find it very hard to believe that ἡ κβ should not be ὁ κβ, in line with the rest of this class of numbers. The arithmetic is straightforward. Could it be a copying error after the preceding η (in a text without diacritics)? Numbers are often copied wrong, but here it’s the article not the number that’s off. Is there an app.crit.?

Thanks for the encouragement, mwh. As hard is it is to believe, all the Internet texts that match the Google search "“ὁ ϛ, ὁ ι, ὁ ιδ, ὁ ιη” indeed have the next number of the series preceded by a feminine definite article! I am still looking for an apparatus to explain that particular passage.
I reread that Nicomachus article. The only “lead” it gives is the dichotomy mentioned between “divine” and “scientific” number. I will try discern a pattern between how numbers were thus referred to and gender/number employed.

https://archive.org/details/autolycidesphaer00autouoft/page/n480/mode/1up

Through the rest of the text though 22 seems masculine.

So, just a random feminine number in that series in sent #51? Copy error propagated in all subsequently copied texts? Would not be the first time, right?

For comparison, here are other examples of feminine numbers. In the first example it cannot be random.
Chapter 8, line 17 has feminine η and ιϛ,

δυνάμεων πρὸς μέρη προχωρήσει τάξει, πρῶτον μὲν ἐπὶ τοὺς παῤ ἑκάτερα δύο, εἶτα ἐπὶ τοὺς ὑπερκειμένους ἑκατέρωθεν, μέχρις ἂν ἐπὶ τοὺς ἀκροτάτους ἀφίκηται, ὥςτε καὶ τὸ ὅλον ἀντιπαρωνυμεῖσθαι τῇ μονάδι καὶ τὴν μονάδα τῷ ὅλῳ· οἷον λόγου χάριν, ἐὰν τὸν ρκη θῶμεν τὸν μέγιστον, ἀρτιογενεῖς ἔσονται αὐτῷ αἱ ἐκθέσεις τῶν ὅρων, ὀκτὼ γὰρ αἱ μέχρις αὐτοῦ πᾶσαι, καὶ μίαν μεσότητα οὐχ ἕξουσιν, ἀδύνατον γὰρ ἐν ἀρτίῳ, ἀλλʼ ἀναγκαίως δύο, > τήν τε η καὶ τὴν ιϛ> , αἵτινες ἀνταποκρινοῦνται ἀλλήλαις παρὰ μέρος· τοῦ γὰρ ὅλου τοῦ ρκη ὄγδοον μέν ἐστι τὰ ιϛ, ἔμπαλιν δὲ ἑκκαιδ έκατον τὰ η· καὶ προιόντες ἐφʼ ἑκάτερον τέταρτον μὲν τὰ λβ, τριακοστόδυον δὲ τὰ δ, καὶ πάλιν ἥμισυ μὲν τὰ ξδ, ἑξηκοστοτέταρτον δὲ τὰ β, καὶ τελευταῖον κατὰ τὰς ἀκρότητας ἑκατοστοεικοστόγδοον μὲν ἡ μονάς, ὅλον δὲ κατὰ τὴν μονάδα ἔμπαλιν τὰ ρκή. ἐὰν δὲ ἐν περισσοῖς ὅροις ἡ ἔκθεσις γένηται, οἷον ἐν ἑπτά, προχειρισαμένων ἡμῶν τὰ ξδ, ἡ μεσότης ἀναγκαίως μία ἔσται κατὰ τὴν τῶν περισσῶν φύσιν καὶ αὐτὴ μὲν ἑαυτῇ ἀνταποκρινεῖται διὰ τὸ σύζυγον μὴ ἔχειν, οἱ δὲ ἑκατέρωθεν αὐτῆς ἀεὶ ἀλλήλοις, μέχρις ἂν εἰς τὰ ἄκρα ἡ ἀνταπόκρισις τελευτήσῃ· οἷον ἑξηκοστοτέταρτον μὲν ἡ μονὰς ἔσται,

Further down, I have feminine κε, με, πα,

πάλιν δὲ ἐκ τῆς τετρακισεπιπέμπτου, οἷον > τῆς κε, με, πα> ,

No, it would not be the first time a copy error propagated to all subsequently copied texts. But this one looks strange, and I wonder if it’s that more marvelous beast: a German typo. All of the digital editions are from the Teubner, I’m sure.

It’s not line 17, but page 17 of the Teubner I linked to, and you are starting in the middle of a sentence.

I’d need to spend some time reading this sort of author to pick up all the words, but in the specific part you are asking about, he is giving a worked example, and is saying that the 8 members of the series generated somehow from 128 have not one μεσότητα (feminine, must mean “median”), but two.

οἷον λόγου χάριν, ἐὰν τὸν ρκη θῶμεν τὸν μέγιστον, ἀρτιογενεῖς ἔσονται αὐτῷ αἱ ἐκθέσεις τῶν ὅρων, ὀκτὼ γὰρ αἱ μέχρις αὐτοῦ πᾶσαι, καὶ μίαν μεσότητα οὐχ ἕξουσιν, ἀδύνατον γὰρ ἐν ἀρτίῳ, ἀλλʼ ἀναγκαίως δύο, τήν τε η καὶ τὴν ιϛ…

128 has 8 factors, of course (1,2,4,8,16,32,64,128), and the middle two are 8 and 16.

Written earlier, before the last two posts. I may come back on this, but I’m out of time right now.

I’m not keen to read this stuff, influential though it was for centuries, but I expect the things that you found puzzling, Mark, are readily explicable, and if Joel were to provide a clear copy of the whole treatise I may skim through and see what I can make of it. The archive copy is fuzzy. Or you could supply examples, as you offered to. ἡ κβ I provisionally take to be a copying slip that should have been noted and corrected, or maybe it’s just a typo.
And In line 18 of the bit that Joel reproduces εχων should obviously be εχον neuter (actually reported as a v.l.) but the editor doesn’t make the correction. The confusion is ubiquitous in early medieval manuscripts. Despite the manuscript data recorded in the app.crit. the editing seems to have been quite mindless. That’s excusable only if the editor aimed to report the manuscript readings and nothing more. Bekker sometimes did that.

P.S. Joel well explains the feminines.

Excluding the majority type of my original post – 1) masculine singular – appositives to arithmos,
My type 2) breaks down as follows:
a) 4x NsGf

Joel correctly cleared the first few up as appositives to mesotēs. It was clear as soon as he pointed it out. Thanks, Joel.
The second example of feminines might have just as simple an explanation (apposition of diplasia?).
Here is the complete sentence (from the last of the book).

ἀλλὰ μὴν καὶ ἑτέρως ἐκ τῶν ἐπιμερῶν οἱ πολλαπλασιεπιμερεῖς καὶ ἑτερογενεῖς ἐπιμερεῖς ἀναφαίνονται, οἷον ἐκ μὲν τῆς δισεπιτρίτου ἀπὸ μὲν τοῦ ἐλάττονος ὅρου ἡ διπλασία καὶ δισεπίτριτος · ἐκ δὲ τοῦ μείζονος ἡ τρισεπίπεμπτος, ὡς ἐκ τοῦ θ, ιε, κε ἤτοι θ, κδ, ξδ ἢ κε, μ, ξδ · ἐκ δὲ τῆς τρισεπιτετάρτου ἐκ μὲν τοῦ μικροτέρου ἡ διπλασία καὶ τρισεπιτέταρτος, ἐκ δὲ τοῦ μείζονος ἡ τετρακισεφέβδομος, ὡς ἐπὶ τοῦ ις, κη, μθ ἤτοι ις, μδ, ρκα ἢ μθ, οζ, ρκα . πάλιν δὲ ἐκ τῆς τετρακισεπιπέμπτου, οἷον τῆς > κε, με, πα> , ἀπο μὲν τοῦ ἐλάσσονος ἡ διπλασία καὶ τετρακισεπίπεμπτος ἐν τοῖς κε, ο, ρϙς, ἀπὸ δὲ τοῦ μείζονος πάλιν ἐπιμερὴς ἢ πεντακισεπένατος ὡς ἐν τοῖς πα, ρξ, ρϙς, καὶ κατὰ τὰ ἑξῆς ἐπ’ ἄπειρον ἀνάλογα καὶ εὐάρμοστα εὑρήσεις ..

b) 9x NsGn

ἐν δὲ περισσαῖς ἐκθέσεσιν ἶσον > τὸ ἅπαξ ξδ τῷ δὶς λβ καὶ τοῦτο τῷ τετράκις ιϛ καὶ τοῦτο πάλιν τῷ ὀκτάκις η > μόνον μέσου πρὸς ἑαυτὸν πολλαπλασιαζομένου …



τὸ δὲ σον > ἐξ ἀντιστροφῆς γ καὶ > τὸ θ-ον β> , κἀπὶ τῶν ἑτέρων ὁ αὐτὸς εὑρεθήσεται τρόπος …



ὁ μὲν γὰρ η μέρος ἔχει ἥμισυ, τέταρτον, ὄγδοον, ἅπερ ἐστὶ δ, β, α, συγκεφαλαιωθέντα δὲ εἰς > τὸ αὐτὸ ζ > γίνονται καὶ ἐλάττονα τοῦ ἐξ ἀρχῆς …



διὰ τοῦτο πολυπλασιάζω αὐτὸν τῇ τοῦ ἐσχάτου προσληφθέντος εἰς τὴν σωρείαν ποσότητι καὶ ἀποβαίνει μοι ὁ κη τοῖς ἰδίοις μέρεσιν ἶσος, ἔχων καὶ αὐτὸς ἐκ τῶν προηγουμένων τὰ ἐν αὐτῷ μέρη, ἥμισυ μὲν παρὰ τὴν δυάδα, τέταρτον δὲ παρὰ τὴν ἑπτάδα, ἕβδομον δὲ > παρὰ τὸ δ> , τεσσαρεσκαιδέκατον δὲ παρὰ τὴν τοῦ ἡμίσους ἀντιδιαστολήν, εἰκοστόγδοον δὲ παρὰ τὴν αὐτοῦ παρωνυμίαν, ἥτις ἐν πᾶσι μονὰς ὑπάρχει …



εὑρημένων δὲ τούτων, ἐν μὲν μονάσι τοῦ ϛ, ἐν δὲ δεκάσι τοῦ κη, εἰς τὴν ἐφεξῆς πλάσιν τὸ αὐτό σε δεῖ ποιῆσαι … ἐὰν μὲν οὖν δὶς μόνον μετρῇ τὸν ἐν συγκρίσει μείζονα, ὑποδιπλάσιος λέγεται ἰδίως, ὥσπερ τ> ὸ α τῶν β> , ἐὰν δὲ τρίς, ὑποτριπλάσιος, ὥσπερ τῶν γ τὸ α, ἐὰν δὲ τετράκις, ὑποτετραπλάσιος, ὥσπερ > τὸ αὐτὸ α τῶν δ> , καὶ ἐφεξῆς οὕτως

My type 3: Numbers refered to with a plural neuter article, or one that could be neuter or masculine, breaks down thus:
a) 2x NpGmn
b) 21x NpGn
…and I will provide examples of these in the next post.

I’m sorry, but I looked at all of the editions on Hathi, and the other copies are all considerably worse.

However, this very fuzzy version of the 1862 printing seems not to have the error. (It’s the 1866 version that the digitization was made from, and which I linked earlier.)

https://babel.hathitrust.org/cgi/pt?id=uiug.30112023810499&view=1up&seq=21

Joel, the hathi trust link you used is new to me. I am relieved to find an edition without that strange feminine in the series!
https://babel.hathitrust.org/cgi/ls?field1=ocr;q1=nicomachus;a=srchls;lmt=ft FTW.

So it was just a typo all along.
I’ll leave it to Joel to explain ἐκ τῆς τετρακισεπιπέμπτου and the rest. It all makes good sense, but you have to go beyond the numerals and try to understand the math.

I rather enjoy this odd classification of even numbers—pariter par (ἀρτιάκις ἄρτιος), pariter inpar (ἀρτιοπέριττος), inpariter par (περιττἀρτιος). I’m familiar with this section from Boethius (largely taken from Nicomachus), which I once had my STEM-inclined Latin students read. It’s not so hard to understand (even for the innumerate like me) once you get accustomed to the technical vocabulary.

In terminology at least, it’s on its way to actual number theory. I was about to type up some uninformed speculation about how useful it would be for talking about prime numbers, but I see flipping through some of the later chapters that we get to the famous “ὑπὸ Ἐρατοσθένους καλεῖται κόσκινον”, making it not speculation after all.