I am looking for comprehensive lists or research papers covering language-specific phonotactic constraints.

It would be great to start with English, but any language will do.

Specifically I'm interested in a nice, packaged set of rules a-la,

These sound combinations are not allowed:
  /bn, dn/
These sound combinations are not allowed at the end:
  /e, æ, ɐ, ʊ, ɔ/

The site provides some nice tidbits and examples, http://clas.mq.edu.au/phonetics/phonology/syllable/syll_phonotactic.html, but I'm wondering if there are more academic or comprehensive papers on this subject.

  • Please ask about a single language, otherwise the question is too broad and I might have to close it.
    – Alenanno
    Jul 4, 2012 at 10:07
  • I specified that I'd prefer English, but stipulated that other languages would be fine. It's not clear to me why that is problematic, given that I'm just looking for references.
    – si28719e
    Jul 4, 2012 at 12:43
  • If you allow any language, you make it too broad. Given that you're asking for references though, there might not be a problem. :) Don't worry, I'm not closing it for now. I'll see the answers it gets, but I don't think there will be a problem.
    – Alenanno
    Jul 4, 2012 at 12:53

2 Answers 2


There is a database in progress hosted at ANU.


I've been wondering the same thing myself, but specifically for Japanese. I'm starting to think that something like a phontactic rule set is only "complete" or "comprehensive" insofar as the level of detail desired. Like how IPA transcriptions can be as narrow as needed, a phonology needs only to be as detailed as the application demands. I found this book The Phonology of Japanese which claims to be "the fullest account of Japanese phonology ever published" all in 320 pages. I ordered the book just yesterday so we'll see. If I wanted a complete guide to American English, however, then I would just get a bunch of books on the subject of "English phonetics and phonology" and compile one myself. It may not, after all, be very large.

  • good point. on Japanese amazon.com/Introduction-Japanese-Phonology-Series-Linguistics/… by Timothy Vance is also very good, if more of an introductory text. I'm interested in combining this expert knowledge with machine learning techniques to see how they might complement each other, so indeed it doesn't have to be complete. If I can't find anything I guess I'll muddle through on my own!
    – si28719e
    Jul 6, 2012 at 2:54
  • @si28719e That's a pretty interesting idea. Are you talking about using attested phonotactics as a metric for evaluating machine performance? Do you have data already?
    – taylor
    Jul 7, 2012 at 22:54
  • @si28719e I was browsing amazon for phonology texts and found this amazon.ca/Learnability-Optimality-Theory-Bruce-Tesar/dp/…. Sounds like it was related to what you were talking about, but I haven't really studied automata yet, so I couldn't really say lolz.
    – taylor
    Jul 11, 2012 at 23:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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