Minding Your Ps and Qs or Why Porcom is a Headache

Today we'll take just a little dip into the history of Gaelic which starts about 5000 BC, so fasten your seatbelts.

The horrible history

5000 BC is roughly when the first Indo-Europeans started invading Europe. We say "invading" because we know people were there before them. Amongst this lovely bunch of hooligans, from the steppes of Central Asia, there was a group which settled on the northern edges of the Alps. The Celts. Back then, they weren't known as the Celts and the two earliest "Celtic cultures", that we know about, are often called the Hallstatt and the La Tène Cultures. Irrespective of the naming issue, this bunch did well and by the 3rd century AD they'd established quite a track record. They muscled the Etruscans out of most of northern Italy, had taken over most of Gaul as well as large swathes of the Iberian peninsula, Southern Germany, the British Isles, parts of modern day Slovenia, Bulgaria, Romania, Hungary, a fair chunk of land in central Turkey(!), sacked Delphi in 279 BC and Rome, itself, in 390 BC.

Incidentally, it's from the Greeks that the Celts got their name. The historian, Hecataeus, described them as Keltoi, the meaning of which can not be absolutely ascertained. But, seeing that they sacked Delphi, it can't have meant anything nice.

However, after that, the Celts slipped a bit. In 192 B.C., Rome took back Transalpina and gradually took over... well ... really most of Europe and the decline of the Celts began.

Indo-European

So what about the language? Patience! The main thing that distinguished the Celts from other Indo-Europeans, in terms of their language, was the loss of Indo-European p. Pardon? Well, Indo-European, which is not recorded, but reconstructed based on what we know of its daughter languages, seems to have had an elaborate system of stops, 12 of them:

p	t	k	kʷ
b	d	g	gʷ
bʰ	dʰ	gʰ	gʷʰ

Celtic

Now we come to Celtic, meaning very old Celtic. Those speakers decided to drop the entire set of aspirated voiced stops and made do with just 8 stops:

p	t	k	kʷ
b	d	g	gʷ

One thing you need to know about that little superscript ʷ is that it represents something called labialisation. It means that you round your lips when making that sound, like in the English word quick which is [kʷɪk]. This is important. Why? You'll see.

Late Common Celtic

Next, for whatever reason, Late Common Celtic dropped p and said k wherever there was a p before. It just does. That leaves us with:

	t	k	kʷ
b	d	g	gʷ

Goidelic and Brythonic

Then things got interesting because this was roundabout the time when Goidelic (the granfer of Irish, Gaelic and Manx) and Brythonic (granma of Welsh, Cornish, Cumbric and Breton) put in for a divorce. Over a p. What happened is that Brythonic took the kʷ sound and turned it into a p. That works because labialisation is made with the lips and there seems to have been a struggle between the labial nature of the ʷ and the velar nature of the k. It appears that the lips won and the k bit was assimilated into a p. It's like the word immigrate which comes from in-migrāre where the n has been assimilated into an m because it is immediately followed by one.

Goidelic on the other hand would have none of it and did not redevelop the p. However, out of sheer spite, it merged the labial series with the plain stops so that kʷ merged with k and gʷ with g.

So, this in Goidelic:

	t	k
b	d	g

But, this in Brythonic:

	t	k	p
b	d	g	gʷ

So what?

This is the reason for a great many things. For example, it is the reason why Goidelic is sometimes referred to as Q-Celtic and Brythonic as P-Celtic. It's based on the development of the Indo-European word for 5, penkʷe which in Q-Celtic becomes cóic (remember, Goidelic dropped p) and in Brythonic pimp (remember, Brythonic kept p). That explains the P but not the Q. Well, it does explain it in Manx because cóig is spelled queig.

So, what else does it explain? It explains why modern Brythonic languages have a gap - meaning there's no historic [kw] sound which explains why they have [p] where modern Goidelic languages have a [k]:

	Indo-European	Loanword from	Gaelic	Irish	Manx	Welsh	Cornish	Breton
five	penkʷe		cóig	cúig	queig	pump	pymp	pemp
head	?		ceann	ceann	qione	pen	penn	penn
sense	?		ciall	ciall	keeaill	pwyll	poell	poell
who	kʷos		có	cé	quoi	pwy	piw	piv
wool/feather	petna	Lat. plūma	clòimh	clúmh	clooie	plufyn	pluvenn	plu

It also explains some lovely loanwords like Càisg for Easter which is derived from ecclesiastical Latin Pascha (cf. Sp. Pasqua) or even the twice borrowed Patricius who shows up as earlier Cadrach and later as Pádraic > Pàdraig. Even some relatively late loans show this change, such as Norse upsi mature saithe which was borrowed as ugsa (and in some areas then metathesized to ucas).

Occasionally this seems to involve a p/g change, like in giseag which also shows up as piseag - the same word as Irish piseog which was borrowed into English as pishogue. This shows up in Middle Irish as piseóc, pisóc with pretty much the same meaning.

It also makes for a headache because modern Irish and Gaelic, as we have just seen, do not retain the kʷ sound but sometimes borrow words from English which has kʷ. How to borrow? Do you borrow the sound kʷ and change the set of sounds in these languages? Or, do you adjust to Irish/Gaelic spelling? Or, do you try to come up with your own word? Tricky one. Traditionally, the second option seems to have prevailed. For example, Irish borrowed Quaker as Caecar and Gaelic turned a quadruped into ceithir-chasach. But, lately, words like quinín 'quinine' have showed up in Irish. and Gaelic now boasts cuaraidh for 'quarry' and cuòta for 'quota'. Really tricky one.

So, what on earth then is porcom? Well, in the Mesopotamian clay table of Ashur-Bannipal... just kidding. There is a 3rd century inscription in Lusitanian, a language spoken in the west of the Iberian peninsula and which is generally described as Celtic, and it goes:

OILAM TREBOPALA INDI PORCOM LAEBO

etc. etc. The tricky bit is that it translates as "a sheep to Trebopala and a pig to Laebo'. And, as we all know, the great clue to something being an old Celtic language is the loss of p before the stage where Brythonic reinvents it. Yet, here we have porcom 'pig' ... The answer? Actually, we don't have an answer except that there may be a question mark over Lusitanian being a Celtic language. If you find out, publish and you'll be famous!

Oh, and in case this kind of thing fascinates you, we've made a little picture for you of what happened to two words - hundred and five - all the way from Indo-European down to over 50 modern Indo-European languages here: