Aut "rex Azteca" aut "rex Aztecus"?

Question

In Spanish, azteca as an adjective doesn't seem to inflect for gender, though it does for number: azteca, aztecas. Hence the phrase el rey azteca.

In Italian, azteco seems to be a perfectly regular adjective.

What precedent is there for preferring one or the other of these in Latin? The etymology says that Spanish azteca comes from a Nahuatl suffix -tecatl. Perhaps people then treated the Nahuatl suffix grammatically by analogy with the Latin/Greek suffix -thēca/θήκη. Coming from classical roots, this suffix only forms nouns, like bibliotheca and discotheca, not adjectives, so the analogy is imperfect, but that's often how it is when extending precedents to new cases.

Another reasonable possibility is to reject the analogy as false to begin with, since the Nahuatl suffix means "inhabitant", whereas the Latin suffix means something more like "container for a collection". But it's also reasonable to think of an ethnic group or civilization as a sort of contained collection.

Answer 1

While many romanizations of ethnonyms take the 2nd declension noun pattern (Germani, Galli, Veneti, Helvetii and so on), the 1st declension, as Joonas noted, also occurs, although I am struggling to come up with any examples beyond Belgae. 3rd declension is also found, but is rare and seems to always be of pre-classical origin.

However, the main reason I am writing this answer is to stress that the Latin idiomatic way of expressing the “king of X” concept is not to name the land, but rather the people over whom the said king reigns, with the noun for the people taking, naturally, the genitive plural. That is, the pattern is not the “German (Swedish, Belgian, American) king,” but instead “the king of the Germans (Swedes, Belgians, Americans).” There are many examples of this very idiomatic pattern, for example, in Caesar [B. G.]: Ariovistus rex Germanorum [1, 31, 10] (masc., 2nd decl.); Teutomatus, Olloviconis filius, rex Nitiobrogum [7, 31, 5] (masc., 3rd decl.); or in Cicero where he acrimoniously mocks Caesar himself as rex populi Romani [Off. 3, 83], “king of the Roman people;” there are many many others. Incidentally, the modern title of the Belgian king is also the “king of the Belgians” (koning der Belgen, roi des Belges).

So, whichever declension of the word for Aztec people you would pick, the idiomatic Latin is to title the king as “King of the Aztecs,” rex Aztecarum/Aztecorum. Personally I would go with the 2nd declension Aztecus, -i as the most common pattern, but this is really your call. Neither way would be wrong.

Answer 2

I don't think -theca is relevant. To me, it seems that Italian azteco and Latin aztecus just represent an analysis where the last "a" in the original Nahuatl word is removed, giving the root aztec-, and then this is treated like the root of a regular first/second adjective, taking the thematic vowel -o- for the stem of the masculine and neuter forms and the thematic vowel -a- for the stem of the feminine forms.

Aztecus is definitely used, but most of the examples I was able to find are in species names, and biological Latin isn't always "correct" from a Classical perspective. Nonetheless, I don't see any strong reason to reject the form aztecus (and the other forms based on the stem azteco-).

It seems that a handful of Latin adjectives did in fact decline, irregularly, as a-stems in the masculine as well as in the feminine: this answer by fvogel cites Forcellini, who mentions ruricola and indigena and says that advena also came to be used this way (and even came to be used as a neuter-gender adjective, apparently).

Answer 3

Your description of the Spanish treatment of "azteca" reminds me of the classical Latin noun Belga, "a Belgian". Belga is masculine, although it looks feminine and follows the first declension. In addition to that noun, there is the perfectly normally behaving adjective Belgicus.

There are similar pairs of nouns and adjectives in English, too, like "Swede" and "Swedish". You can say "Swedish king", but "Swede king" doesn't quite seem to work. This kind of juxtaposition of two nouns works even worse in Latin in my opinion.

It sounds reasonable to treat Azteca just like Belga in Latin — but I would not be opposed to the separate nouns Azteca and Aztecus either. When used together with rex, what you need is not a noun but an adjective. A masculine adjective has to look masculine¹, so rex Azteca sounds definitely wrong to me.

What you have to decide is what to take as the adjective for Aztecs. As sumelic mentions, the adjective Aztecus is already in use, so it is definitely a good choice. You could also use Aztecicus if you want; there are plenty of such adjectives in Latin formed with -icus, including Belgicus. I find myself preferring -icus over -us, but that is a personal opinion. Either way, my suggestion is Rex Aztecus/Aztecicus.

I think comparison with -theca is not fruitful here. Instead, I would simply take the stem Aztec- as given. This stem was formed when the Nahuatl word was reanalyzed and conformed to Spanish, and I see no compelling reason to choose otherwise in Latin.

---

¹ In Latin masculine nouns can look feminine or vice versa. Spanish is more flexible in that respect; the adjective "azteca" can be used for either gender in the same form. In Latin this is impossible in the first declension.

Answer 4

Latin, as a general rule, has two different types of words. The first type inflect fully, and for adjectives that means a full set of case, number, and gender forms. The second type don't inflect at all and use the same form for every case (usually these are only foreign names). And as far as I know, every adjective in the language falls into the first category: there's no adjective which doesn't have a full case-number-gender inflection.

So while the noun "Aztec person" might be borrowed as Astecat (undeclined) or Asteca (first declension), the adjective would have to decline in some way. Either Astecus -a -um or Astecicus -a -um would make sense, though I'd go with the first (back-formed from the noun Asteca) to avoid the pattern -ecic- in the middle.

EDIT: Brianpck has pointed out the adjective nēquam "worthless" is, in fact, indeclinable. I stand corrected! But I believe my point is still valid for foreign/borrowed adjectives.