I've been looking for my answer but I still haven't found what I want. I have a question regarding lenis voiceless+fortis voiceless clusters (in American English). Does the fortis voiceless consonant get assimilated to a lenis voiceless? For example:

gʊːd̥ fɹɛnd - good friend / weɪːv̥ fɔɹm - waveform / meɪːd̥ sʌːm - made some / hæːz̥ sʌːm - has some /

changes to:

gʊːd̥ f̥ɹɛnd / weɪːv̥ f̥ɔɹm / meɪːd̥ s̥ʌːm / hæːz̥ s̥ʌːm /

(I know that f̥/s̥ are not the right symbols to a LENIS voiceless, but what I am implying here, is that both f and s (FORTIS) became Lenis)

So, does a lenis voiceless+fortis voiceless cluster get assimilated? Other possibility would be an assimilation as follows: fortis voiceless+fortis voiceless, but I'm almost sure that's not the case. If there's no assimilation, and both consonants keep their characteristics, lenis followed by a fortis one, is there any tip on how to pronounce such a cluster keeping their characteristics?

Thank you very much!

  • You might get more answers in the linguistics stack overflow – BleepBloopOverflow Jan 20 '16 at 2:03
  • @BleepBloopOverflow Absolutely not. This is the website for questions about the English language for professional linguists. Language-specific questions are off-topic on linguistics SE. For basic questions about English people can go to ELL. – Araucaria - Not here any more. Jan 20 '16 at 2:25
  • 1
    I would say that they do not. It seems to me that assimilation in American English is always towards voicelessness, not towards voicing (with the exception of intervocalic /t/ voicing). I haven't heard of assimilation towards a lenis or fortis state. – herisson Jan 20 '16 at 2:29
  • @sumelic You haven't heard of lenis sounds becoming devoiced when next to voiceless sounds? – Araucaria - Not here any more. Jan 20 '16 at 2:32
  • @Araucaria: I have; that's why I said assimilation towards voicelessness does occur. To clarify what I mean: lenis sounds are often voiced in English, but due to assimilation, they are usually unvoiced next to voiceless consonants. In either case, they remain lenis, though. And as far as I know, fortis consonants never turn voiced or lenis through assimilation. – herisson Jan 20 '16 at 2:38

A fortis consonant is a consonant which we think of as typically voiceless. A lenis consonant is a consonant that we think of as typically voiced.

In general in English lenis consonants become devoiced when not surrounded on both sides by voiced sounds. So in the string bed time the /d/ will tend to be devoiced, either partially or fully, because it has a voiceless /t/ following it. Similarly the /b/ at the beginning of bed time will also become partially devoiced at the beginning, because it is not preceded by a voiced sound.

It is rare for fortis consonants to become voiced, but they do in some instances. So for example intervocalic /t/ will become a voiced tap in most varieties of American English before an unstressed vowel. The fricative, /h/, may also become voiced in some words when it's intervocalic, for example in the word ahead.

In response to the Original Poster's question, no. The lenis consonant at the end of the word will be assimilated to the voiceless sound preceding it or following it. That is the only reason it is devoiced in the first place. Voicing involves the vibration of the vocal folds. Our vocal folds are late to kick in and early to finish in most situations. They're a bit lazy. So we only get a devoiced lenis consonant when it is next to a voiceless sound. In terms of voicing, there is nothing for a fortis consonant to assimilate from a neighbouring sound which also has no vocal fold vibration.

  • Thanks for the response, but what I meant was: both sounds being voiceless, is there any possibility for one sound to change its "energy" from fortis to lenis, being both voiceless? I aware about the VOICING assimilation ^^, thanks for reinforcing it, but that's what I meant: not VOICING ASSIMILATION, but ENERGY ASSIMILATION. – Daniel Jack Jan 20 '16 at 3:06
  • 1
    @DanielJack The notional difference in energy is not a defining characteristic of fortis and lenis consonants in any meaningful way, especially if those consonants are surrounded by other consonants. For example a devoiced /b/ followed by a /p/ will have no release phase where some notional difference in energy might be demonstrated. There is no energy during silence. You may have noticed that there is no diacritic in the IPA for 'energy'. – Araucaria - Not here any more. Jan 20 '16 at 9:00
  • So, for example, webpage, waveform and has some, will be pronounced as: wɛːppeɪːd͡ʒ / weɪːffɔɹːm / hæːssʌːm ? – Daniel Jack Jan 20 '16 at 14:30
  • @DanielJack Good question. The answer is: not quite, because fortis and lenis consonants have diffferent effects on surrounding sounds. So for example, a partially or fully devoiced /b/ will have no aspiration when occurring before a vowel and won't cause devoicing of a following approximant. Also a devoiced lenis consonant won't cause the preceding vowel to be clipped (it won't cause prefortis clipping). – Araucaria - Not here any more. Jan 20 '16 at 14:38
  • Daniel included the lack of clipping in the transcriptions (the vowels have a length mark). @DanielJack: in your transcription, you left out the aspiration on the [pʰ] in "page." Also, the final d͡ʒ is likely to be at least partially devoiced; you should transcribe it as d̥͡ʒ̥ or t͡ʃ. More like [wɛːppʰeɪːt͡ʃ.] – herisson Jan 20 '16 at 16:33

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.