"They will be leaving soon."
Translation:Bidh iad a' falbh a dh'aithghearr.
There is no agreed explanation. Lots of languages have consonants that are 'put in before a vowel to make it easier to say' (English an, Gaelic ag, French donne-t-il, Welsh yr ) but that does not explain why certain ones are used in certain places. Almost always it can be traced back to a consonant that was there originally but dropped when easier to say without.
It is thought that the dh' in dh'fhalbh e is the same as the do in cha do rinn. It marked a perfect tense and later just the past. If so then cha do dh'fhalbh is a bit ridiculous.
But a dh'aithghearr is not so obvious. Mark says
dh’, a dh’ – abbr form of the prep de of
but I do not know if this is generally accepted.
As for the rest, no one seems to know. Aith might be a variant of ath 'next' and geàrr means 'short' (adj.) or 'cut' (verb). But how we assemble these parts to make 'soon' is not clear. D
Check out the Wictionary entry here: https://en.m.wiktionary.org/wiki/aithghearr
And the Old Irish at "aithgerr" (with some neat citations) : http://edil.qub.ac.uk/2666
The prefix ath- is parsed as an intensifier; geàrr "short" perhaps in a similar semantic development with English "short [size] > short [time]". Dh' (=do, the adverbial particle, analogous to gu) might be lenited by assimilation with aith- although I haven't heard of consonant assimilation rules in ScG before this.
Interesting. Your idea about the "short" is sensible but I am not too convinced about ath as an intensifier. There are so few examples, and in particular the Wikionary article that claims this links to the word ath- that does not seem to have this meaning. I am tempted to think they made this up to explain aithghearr, whilst using the same word as evidence that it used as an intensifier.
While checking in various dictionaries, I did come across this in FGB which looks a bit interesting
atha, f. (gs. ~). Space of time. I gceann ~, after a while. ~ fhada, a long while. Le h~, for some time.
That also makes sense as then * atha ghearr is just the opposite of i gceann atha, perhaps with the preposition elided, meaning 'in a short while'. Although atha is also a bit obscure, they do at least quote three compounds that add a bit of weight. The Wiktionary entry for this word takes us to OI athad which I found in eDIL with quite a few citations, generally in adverbials of time. So I think this looks much more promising. D
Some people refer to it as a mutation, for want of a better explanation, but no one has any good suggestion for what it might be a mutation of (except as I discuss below), so I think we have to put that in the category of 'folk etymology'. I discuss one possibility above, that could explain it but only in one circumstance - the past tense. Since we know that that do was used as a past marker, and there is no reason to assume that it was not once used on positive statements as well as questions and negatives, it does seem likely that it was simply abandoned before a consonant but retained before a vowel. (Incidentally, the a does not make any difference - it would be dh'fhalbh anyway).
So the only issue is why it is dh' not d' as it is in Irish (d’imigh). And basically no one knows. There is actually a further problem, and that is that Irish does not usually use do as a past marker, so that casts a bit of doubt on the explanation, without providing a better alternative. So we don't know where it comes from in Gaelic. Then there is also a suspiciously similar thing in Welsh - they say naddo (which would be * nadho in Gaelic spelling, with the lenition presumably caused by the na) where we might say something like Cha do rinn (i.e. meaning 'no' in the past). The dictionary of Old Irish shows this used occasionally in the positive and more often in the negative. There is no doubt where these words come from, because that last link says clearly that th Irish comes from the Welsh, and GPC (link does not work in all browsers) (see under do meaning 1) says clearly that the Welsh comes from the Irish. Very helpful.
So all the evidence points to there having been a preverbal particle in some Celtic language at some time that meant something, probably a past marker.
That then leaves the use in things like a dh'aithghearr. No one has a better suggestion than that it simply got adopted as a way to make things easier to say, and then spread. D
I believe it is thought that ro-, related to English pre- (with loss of p in Proto-Celtic), was a perfect marker, used specifically to distinguish between the preterite and the perfect. For example Srachan gives (pp.64-65) cechain 'she sang' but ro cechain 'she has sung' (my translations). Over time this contrast was lost and ro- was used when convenient in what became a general-purpose past. This general-purpose tense was often called the preterite, but that is very unfair when it contains a specific non-preterite marker. I am not that good at Modern Irish, but certainly in Gaelic, our past tense is often used where you would use the perfect in English, with the compound perfect only being used when it is necessary to emphasise the perfect aspect.
Thus far I think things are fairly non-controversial, but then we get to do-. The usual story is that ro- changed into do- somehow. Meaning they don't know how. Confusion between the two prefixes (which were both common anyway) is suggested, but that is like suggesting that re- and de- would become confused in English, just because we have verbs like receive and deceive, revise and devise, etc. I don't buy it. I think that it was a separate marker, possibly general past or possibly preterite. It predominance in Gaelic and its presence in Welsh suggests that it may be British or from Britain in origin. It might even be pre-Indo-European, in my opinion.
But wherever it came from we can agree that is is found as a past marker. So was it preferentially retained before vowels? Did it then come to be thought of as a separate epenthetic d, allowing its use in cha do dh'fhalbh and a dh'aithghearr? Then how did it lenite? Perhaps * d'fhalbh got treated like other past tenses and lenited by levelling, and this then spread to other uses such as a dh'aithghearr. Who knows? D
Modern dictionaries (AFB and Mark) give only 'early' for (gu) moch. I have never heard it in any other sense, although Dwelly does give 'soon' as the third meaning. He gives two examples for 'early' and none for 'soon' and he does not include the gu.
So stick to 'soon' for a dh'aithghearr and 'early' for (gu) moch. D
Unfortunately there is no way to replay that sentence, but I can tell you what is usual for the two occurrences of dh in that sentence.
When it is slender (next to an e or i, as in bidh) it is a /j/ (like a y), although this sound may not be noticed in this word, next to the i.
But when broad (next to an a, o or u, as in dh'aithghear) it is /ɣ/ (the 'rough g' that you describe).
So the pronunciation of the dh' will depend on what word or is attached to, and may well vary by dialect as well.