I recently came across the concept of "svarabhakti" in the context of the Gaelic word "Alba" (pronounced with a vowel between l and b). What I'm confused about is why this is considered a svarabhakti rather than the word simply being "Alaba"?


"Svarabhakti" is from a Sanskrit grammatical terms referring to a rule of pronunciation promulgated in the Prātiçākhyas that inserts a "brief" vowel between r and a nasal, see Whitney's Sanskrit Grammar, 230(c). In contemporary terms, this is not a full vowel, this is a vowel-like concomitant of /r/ in a certain context, part of phonetic implementation. Sometimes such vowel-like sounds are called "epenthetic" vowels, if the vowel that is inserted is the same as some phonemic vowel (apparently not the case with Sanskrit). Sometimes, such articulatorily-driven vowels are called "excrescent" vowels.

There isn't a particularly good reason to use the word "svarabhakti", but there might be a real linguistic question at stake. There is discussion of this issue in Bosch & De John 1997. One question is whether what is inserted is a full vowel as opposed to being a simple "vocalic release". IMO the evidence is good enough that it is not a "mere release gesture". The epenthetic vowel is an exceptionally-stressed non-initial vowel (any vowel quality can be inserted, whereas unstressed vowels are drawn from a restricted inventory). The fact that intervening consonant palatality determines frontness argues that there is a phonological process (this is not phonetic implementation), as does the fact that the inserted vowel is phonetically not a micro-vowel of e.g. 40 msc, it is a full vowel.

However, that does not address the question of claiming that the underlying and surface forms are different: why not just say that [dɔrɔγ] is /dɔrɔγ/ rather than /dɔrγ/. The rationale is that there are phonetic details that force such an analysis, e.g. that with non-epenthetic /a/ an intervocalic consonant is syllabified in the coda ([ar.an] "bread") but with an epenthetic vowel the intervocalic consonant is in the onset ([a.ram] "army"). This is not a particularly compelling argument. Bosch has another paper, "The syllable in Scottish Gaelic dialect studies", Scottish Gaelic Studies XVIII, 1-22, which is said to go deeper into the question. It is important to note (as Bosch does) that not all Gaelic dialects are identical: she is reporting a famous fact from Barra Gaelic, which crucially depends on a particular interpretation of a transcription.


Svarabhakti, or anaptyxis, or, in more common sense, epenthesis can't be different from the vowel because of the different nature of them: a vowel is the vowel, but svarabhakti is the process of the insertion of vowels into clusters of consonants. Also svarabhakti is related with the phonotactics: the word 'Alba' can't be simple 'Alaba' because Contemporary Gaelic 'Alba' /'aLVBa/ (L, B - different allophones, V - inserted vowel) comes from Old Irish 'Albu' /'albu/ where there no need epenthesis too much. And also svarabhakti is related with the orthography: before the recognizing of such type of vowels, they comes from sporadic processes of pronunciation, but then the pronunciation can be influenced by the new spelling, i.e. 'Alaba'.

  • 1
    I don't find any of this answer persuasive. The arguments seem to be 1) it's different because there's a process with a name; 2) it's different because Old Irish; and 3) it's different because it doesn't appear in the spelling. It mentions phonotactics but then proceeds to ignore it. – Colin Fine Apr 6 at 16:51
  • This is my fault, I did type "considered", but I mean "related". – T1nts Apr 7 at 2:11
  • And first part was as answer to the topic's headline. Question was incorrect because of the different nature of the 'svarabhakti' and 'vowel'. – T1nts Apr 7 at 2:14
  • 1
    Then I tried to answer to the question inside: why inserted vowel in the 'Alba' is considered as svarabhakti, not just it is crypto-'Alaba'-word: because of the history of the word and phonotactics of the languages - firstly it was 'Albu' in the Old Irish, where it was easily pronounced without any epenthetic vowel, or only some people inserted it. Then it becomes 'Alba', and more peoples start to insert the vowels. The different vowels in the different languages: a or ə. This is related to phonotactics of the language. But also we have the literary tradition, where word is still 'Alba'. – T1nts Apr 7 at 2:29
  • So until there is 'Alba' spelling in the languages, we say that there is process of inserting of the epenthetic vowel. And even we would write it as 'Alaba' some people would say it as 'Alba', there is still the process inside. But when most of the people become to pronounce it as 'Alaba' it seems to be new statement. – T1nts Apr 7 at 2:36

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.