"Feumaidh tu tiocaid."
Translation:You need a ticket.
In this sentence, the pronoun is tu. In sentences where a form of to be comes first, the pronoun is thu.
That reminds me of consonant mutations in Welsh.
Is it the case that initial verbs that are forms of to be cause h-prosthesis of the following subject pronoun (or another mutation I don't know the terminology for), but other verbs don't?
You are confusing a few things here.
- This is not the h-prothesis (or, as you wrote, h-prosthesis) – that happens when the sound /h/ appears before words beginning with a vowel, eg. na h-Alba of Scotland with the prothetic h before Alba.
- The mutation in question here is lenition (the only consonant mutation that really happens in Gaelic).
- It is not caused by the verb. In Scottish Gaelic the pronoun tú shortened its vowel and got lenited basically everywhere, becoming modern thu – that’s the default form (historically tú could be lenited in various contexts, and modern Scottish Gaelic generalized the lenited form as the default). See how eg. James is there would be tha Seumas ann – no lenition to Seumas.
Now, the question is: why thu is not lenited here?
And the answer is that thu, tu, because it starts with a historically dental stop /t/, is delenited (or resists lenition) after verb forms historically ending in a dental consonant (one of: D, N, T, L, S; or historically lenited DH, TH) – because in those contexts, when both forms (lenited thu and not lenited tu) still coexisted, the lenited one could never happen.
Since feumaidh ends in -idh which historically was dental /ð’/, in that context thu is delenited to tu. This will happen mostly in future tense and conditional mood which end in -idh and -dh respectively. But also after past copula bu (which historically was budh), eg. bu tu an gaisgeach what a hero you were!, lit. you were the hero.
See also Personal pronouns in Wikipedia article on Gaelic grammar.
1) Thanks for this. Sorry for calling it h-proSthesis. I'd love to say that was a typo but I think I just misunderstood the term!
2) So it is a mutation, and it's lenition. I've got another question if I may: I thought lenition was about palatisation. But neither tu or thu are palatised as far as I can tell. What exactly is lenition then?
3) I'm a bit confused. You say that historically tù was lenited to thu. Then you say "Now, the question is: why thu is not lenited here?" Does this mean that lenition of a lenited consonant doesn't take you back to where you were? tù → thu → tu ?
Your main explanation is crystal clear! Thank you hugely, that really helps. It also helps explain one reason why Gàidhlig spelling can seem redundant to me - in at least this case, it can tell you when or when not to lenite, by showing the historical phonological environment.
1) proSthesis is OK too, both names are used, see here ;-).
2) Lenition comes from Latin lenis ‘soft, weak’. It means weakening consonants, making them more like vowels (more sonorous on the sonority scale). The strongest consonants are voiceless stops – they stop the airflow during speech entirely and are as farther from being a vowel as possible – there is no possibility to have a voiceless stop being a nucleus of a syllable; the weakest sounds in their consonantality are vowels which are voiced and don’t block the airflow at all; in the middle you have fricatives and liquids (or sonants) like l and r (which generally work as consonants, but can be the nuclei of syllables in some languages, see eg. Czech krk ‘neck’, one syllable).
Now, this weakening in Gaelic happens by:
- changing most stops (/p/, /b/, /k/, /g/) and nasal /m/ to fricatives with the same place of articulation (respectively: /f/, /v/, /x/, /ɣ/, and again /v/, historically through nasal /ṽ/),
- /t/ and /d/ to respectively /h/ and /ɣ/ (so dh and gh are pronounced the same today), but historically in Old Irish it also was the fricatives with the same place of articulation: /θ/ and /ð/,
- /s/ to /h/,
- and removing /f/ completely,
- also liquids /r/, /l/, /n/ are affected to some extent but not marked in writing (slender /L’/ [ʎ̪] to [l̪], the trill /R/ [r] to a tap [ɾ], both broad [n̪ˠ] and slender [ɲ] to [n]).
In Welsh lenition (or soft mutation) does similar thing to voiced stops (they turn into voiced fricatives, /b/ → /v/, /d/ → /ð/, but /g/ disappears), but voiceless stops are voiced instead (eg. /p/ → /b/).
This is a quite common phonological phenomenon in many languages, eg. it’s very similar in Spanish where /b/ between vowels is pronounced [v] (or actually bilabial [β]) and /g/ between vowels is pronounced as [ɣ]. But in most languages it is purely a phonological thing where the consonants are weakend between vowels. In Celtic languages original vowels causing the weakening disappeared, but the mutations stayed as a grammatical feature.
3) Em, no. Lenition weakens the consonant, it cannot make it stronger again. ;-)