In the Nahuatl language, the word "altepetl", which means a city-state, comes from the combination of "atl" and "tepetl". But according to the word combination rule in the Nahuatl grammar, the suffix -tl (and -tli) of the preceding word would be omitted (such as tepetl + yacac = tepeyacac). Why doesn't the word "altepetl" obey this rule (being "altepetl" instead of "atepetl")?

EDIT: Here is my speculation: Maybe the word "altepetl" was one of the early words that Spanish heard from the Nahuas (think about how they talked about cities at the very beginning of the contact) and possibly at this time it was actually spelled as "atepetl". But because it was the very beginning of the Spanish-Nahuas contact, the Spanish was less careful about the correct transliteration of Nahuatl words resulting in the misspelled "altepetl". The misspelled "altepetl" then learned by the subsequent Spanish, and then entered dictionaries and grammars textbook written by Spanish scholars without correction. Not so long after the conquest, reversely, the Spanish taught Nahuatl to both local and remote indigenous peoples so it ended up that the spelling "altepetl" is preserved through generations and no one notice it. Can my theory be true?


The word was probably not an actual compound, rather it was a conventionalized collocation, similar to English "kick the bucket". Andrews in his Introduction to Classical Nahuatl indicates that the absolutive marker -tl is not used in the first member of a compound, so nene-tl + pil-linenepilli. Karttunen in her dictionary comments for the specific word āltepētl that "[w]hen possessed, the elements of the compound generally separate, with the possessive prefix attaching to each one", and (under tepē-tl) notes the form totāuh totepēuh, where each noun bears case separate case and possessive marking. Since it's not actually a compound, the elements would each retain their case marking.

The change ātl tepētlāl tepētl does not seem to represent a general phonological rule per se, since /tˡt/ clusters do not seem to arise word-internally, but it it not a surprising phonological change, in that is eliminates a difficult-to-parse elements from syllable-final segment.

| improve this answer | |
  • 1
    Are there any instances of tlt in the language? I can't think of any off the top of my head. – Draconis Nov 28 '18 at 0:41
  • As far as I know, I have never seen any word with the "tl → l" transformation in the language besides "altepetl". Could you give other examples? – armamoyl Nov 28 '18 at 3:42
  • 1
    Andrews gives /mīl-tlah/ → [mīllah] 'place where there is an abundance of cultivated land'. – user6726 Nov 28 '18 at 5:51
  • I don't think there are any instances of tlt in Classical Nahuatl. That's because, with the exception of the absolutive suffix, tl may only occur in a syllable's onset – OmarL Dec 1 '18 at 19:39
  • The question is, are there any contexts where underlying /tl+t/ can be constructed (syntactically or morphologically) and is then changed? – user6726 Dec 1 '18 at 20:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.