Strange article in the UK Independent

markesmith

Link from Independent (UK)

Reddit discussion that clarifies it a bit

I'm not sure what to make of this article. The title is pure clickbait, and the text is a bit jumbled.

Seems to be saying that the DNA from bones found under a pub in Rathlin Island correspond strongly to modern Irish DNA, even though it's from the Beaker or Bronze Age. It's refuting the (to my mind disproven) 'Celtic origin' theory.

Any thoughts on this? Between this and the Palaeolithic find, a big week for Irish archaeology?

Wibbs

IMH the second we start to talk of "Celts" we're down the rabbit hole in many ways. Even among the Romans and Greeks who lived at the same time as the Keltoi couldn't make up their minds where they were from or if they had a consistent culture. Their general position was they were people that weren't us. A bit exotic. Out there somewhere.

Yes Ireland shows evidence of broadly "celtic culture", but this doesn't mean that actual central European peoples showed up here. Christianity got here not so long after and really influenced the culture but that didn't make us Italian, Greek or Jewish.

markesmith

What's current thinking with DNA? Are we mainly descended from Mesolithic, Neolithic, Beaker or Bronze Age? It all seems quite confusing, I've yet to hear a definitive answer on it.

dubhthach

markesmith wrote: »

What's current thinking with DNA? Are we mainly descended from Mesolithic, Neolithic, Beaker or Bronze Age? It all seems quite confusing, I've yet to hear a definitive answer on it.

All three, plus we now know that part of Steppe ancestry is what could be termed "Caucasus Hunter Gather"



What the Rathlin remains show is that the Bronze age marks a turning point in population history of Ireland. The remains from Down (Neolithic) in comparison to both Rathlin remains (500 years later) and modern Irish, actually resemble populations from Mediterranean more.

In simplistic terms one could compare the process of admixture that has happened in South America over last 500 years to what has happened in Europe over the last 8 thousand years (or more, depending on when Neolithic started in an area).

The Rathlin island remains show high levels of simalirity to modern Irish, however modern Irish have more "Neolithic admixture", this makes sense as the Rathlin type men would have intermixed with whatever neolithic population in Ireland producing a mixed population.

Just as many Mexican's today have Amerindian ancestry, over thousands of years you end up with a "blender effect" where everyone ends up with about equivalent levels of admixture.

By and large on a three way admixture analysis you could say following for modern Irish:
1. Pluarity of genome contriubution from Mesolithic population (who by way probably had dark skin but blue eyes)
2. Followed by Neolithic (not too far behind -- shift to paler skin probably due to agriculture and deficiency in Vitamin D in grain based diet)
3. Steppe type ancestry that arrives during Bronze age

the "Steppe type ancestry" is basically seen in all modern populations who speak a Indo-European language (or in non-IE speaking populations living beside IE speakers), it's associated with aDNA taken from Yamnaya and Corded Ware remains.

So Rathlin men probably represent advance of some dialect of Proto-IE into Northwestern Europe. The timeframe probably pre-dates the division of western Proto-IE into Italic, Celtic and Germanic branches. However these men would have probably spoken language that was fairly closely related to Proto-Celtic, as result if dialects of Proto-Celtic were to spread during later Bronze age (trading networks of Atlantic Bronze age for example) than shifting language wouldn't be that hard.

The question arises is where did Proto-Celtic arise within region of Proto-IE in western Eurasia and what was mechanism that led to it's spread (contact networks, with elite emulation? etc.)

dubhthach

I should note that the Rathlin men own neolithic ancestry shows more affinities with neolithic remains from modern Germany than with the neolithic remains from Down.

dubhthach

It's worth reading the original paper which is available on PNAS see:

http://www.pnas.org/content/113/2/368.full.pdf?with-ds=yes

slowburner

Quick question: the Irish melting pot, did it ever include a brown eyed trait from any prehistoric populations/migrations?

dubhthach

slowburner wrote: »

Quick question: the Irish melting pot, did it ever include a brown eyed trait from any prehistoric populations/migrations?

Well the Neolithic woman from Down was brown eyed, I could be wrong but I think the three Rathlin men might have been brown eyed as well (would need to check).

The beauty of newer techniques is that we are now getting full genomes, so as a result you basically get the full "library" whereas before previous methods were basically just looking at "paragraph's" from one single book -- to stretch my library analogy

dubhthach

As an aside they are doing work with aDNA with this massive Bronze age battle site in Northern Germany

http://www.sciencemag.org/news/2016/03/slaughter-bridge-uncovering-colossal-bronze-age-battle?utm_source=sciencemagazine&utm_medium=facebook-text&utm_campaign=bronzeagebattle-3174

http://gepris.dfg.de/gepris/OCTOPUS?context=projekt&id=236942581&language=en&task=showDetail

a minimum of 130 individuals retrieve from just one small part of site.

dubhthach

slowburner wrote: »

Quick question: the Irish melting pot, did it ever include a brown eyed trait from any prehistoric populations/migrations?

From paper

Phenotypic Analysis. Ireland is unusual in displaying world maximum
frequencies of a number of important genetic variants,
particularly those involved in lactase persistence and two recessive
diseases: cystic fibrosis and hemochromatosis (38–41, 42,
43). Interestingly, both the high coverage Neolithic and Bronze
Age individuals were heterozygous for the hemochromatosis alleles
H63D and C282Y, respectively. Modern Irish allele frequencies
are 15% and 11% for these variants, with the latter,
more penetrant variant responsible for a world maximum of this
disease in Ireland (44, 45). Additionally and in accordance with
other data suggesting a late spread of the lactase persistence
phenotype now prevalent in western Europe (7, 10), the Neolithic
Ballynahatty was homozygous for the nonpersistent genotype
and Rathlin1 was heterozygous and thus tolerant of drinking
raw milk into adulthood.
We were able to deduce that Neolithic Ballynahatty had a dark
hair shade (99.5% probability), most likely black (86.1% probability),
and brown eyes (97.3% probability) (46). Bronze Age Rathlin1
probably had a light hair shade (61.4%) and brown eyes (64.3%).
However, each Rathlin genome possessed indication of at least one
copy of a haplotype associated with blue eye color in the HERC2/
OCA2 region.

So probable the Rathlin men were brown eyed but they were carriers for variant associated with blue eyes (in scenario where they had children with blue eye woman there would be 50% chance children would be blue eyed). Adult lactase peristance (ability to drink milk as an adult) appears to have spread into western Eurasia from Steppe where it arose among nomadic pastoralist's, the earliest signs of it are in Yamnaya remains, where also the oldest trace of the Plague has been found.

What's interesting about this Bronze age Plague sample is that it lacks genes to survive in Flea's, ergo the original form of Plague was "pneumonic plague", with "bubonic plague" only evolving later after acquiring of ability of plague bacteria to survive in flea's.

markesmith

Fascinating stuff! Thanks as always dubhtach.

What interests me about this field is how quickly theories rise and fall. Almost as soon as Oppenheimer's book came out, it was largely discredited. It's very much an area in its infancy.

dubhthach

markesmith wrote: »

Fascinating stuff! Thanks as always dubhtach.

What interests me about this field is how quickly theories rise and fall. Almost as soon as Oppenheimer's book came out, it was largely discredited. It's very much an area in its infancy.

Well it took a couple of years to discredit Oppenheimer, however unfortunatley due to TV shows/books etc some of ideas have stuck with general population.

Part of reason for this is that the early genetic studies were so "under-powered" you have to remember the first full Human genome cost something like $3billion (for one genome) and was only completed at about the same time.

Thus early studies only looked at very small segments of DNA, and than tried to extrapolate these. So for example Haplogroup R looked like this in 2003:



Oppenehimer and the like honed in on R1b-M269, which has you can see in 2003 they had a number of branches to (including R1b-M222). The problem is that they were missing all the intermediate branches/nodes.
So they had a branching structure like this:
M269 -> M222

By 2009 the tree looked like this:

so for R1b-M222 men the "cookie trail" now looked like this:
M269 -> L23 -> P310 -> P312 -> L21 -> M222

Today in 2016 the "cookie trail" for M222 looks like this:
M269 -> L23 -> L51 -> L151 -> P312 -> L21 -> DF13 -> DF49 -> Z2980 -> Z2976 -> DF23 -> Z2961 -> M222

Likewise M222 has been broken into multiple branches such as in this tree, some of which appear to have geographic/surname bias (for example A259 appears linked to surnames of Uí Briúin origin in Connact)



Now a days it costs a couple grand to do a full genome (circa $2-3,000) and there is every growing number of them available (project such as "1000 genomes project" etc.)

That and we now have ancient DNA, whereas other studies were based purely on modern samples, and in some ways impart biases from researchers (the ideas that the Basques were a a Paleolithic remnant population for example)

dubhthach

Of course even before aDNA blew away some of those ideas the writing was on the wall based on increasingly large studies which showed that for example in western Europe that R1b was lacking diversity and seem to be dominated by large "younger clades", based on distrubution in modern populations (as well as initial aDNA results) you could come up with a map like this back in 2013



Of course now we have found R1b in Yamnaya remains from Eurasian steppe, as well as seen results from Anatolian neolithic, so we know that map is wrong in having expansion of R1b coming out of Anatolia.

markesmith

In light of this new information, what's your take on the entrance of the 'Celtic' languages to the British Isles and Gaul / Iberia?

Mallory reckons the Gaelic language entered Ireland in the late Bronze Age and was strengthened through the Iron Age, pointing to four key tech developments (hillforts, iron metallurgy, La Tene, refugees fleeing Rome).

I've always had a hunch that the Celtic language (in some form) was in situ before this, maybe even going back to Beaker period. Thoughts?

dubhthach

markesmith wrote: »

In light of this new information, what's your take on the entrance of the 'Celtic' languages to the British Isles and Gaul / Iberia?

Mallory reckons the Gaelic language entered Ireland in the late Bronze Age and was strengthened through the Iron Age, pointing to four key tech developments (hillforts, iron metallurgy, La Tene, refugees fleeing Rome).

I've always had a hunch that the Celtic language (in some form) was in situ before this, maybe even going back to Beaker period. Thoughts?

Well part of nub in this case is when did the sound changes that differenate Proto-Celtic from Proto-Indo-European (or perhaps intermediate Proto-Italo-Celtic stage) occur.

If we are to believe the hypothesis that Reich put forward that aDNA is showing mass input from Eurasian Steppe during Bronze age, and that this genetic input is tied to spread of Proto-Indo-European.

Than it's quite possible that dialects of Proto-IE were in Ireland during the Bronze age (Rathlin island men for example).

The question than arises is when did Proto-Celtic (the common ancestor stage of all known Celtic languages) occur. One of defining features of Proto-Celtic is the loss/deletion of Proto-IE *p

For example compare:
iasc with piscis and fish.

All three words are cognates sharing common descent from reconstructed Proto-IE: *pisḱ-

In Proto-Germanic *p -> f
In Proto-Celtic *p -> ɸ

There's some debate that this may have occurred due to contact with languages which lack *p phoneme such as Iberian (spoken by historic Iberians of eastern Spain) and Basque. Which would point to an origin potentially in south of what is now France.

The question arises did the spread of such sound changes in say Bronze age have a tie in with stuff like trade/contact networks. For example the "Atlantic Bronze age" and "Urnfield" material cultures.

What seems to be case is that oldest known written Celtic language appears to be connected to Golasecca culture of Northern Italy. This is development from earlier Urnfield (in parallel to Hallstat).

Peter Schrijver for example talks about "North Celtic" vs. "South Celtic", within "North Celtic" he groups Gaulish, Brythonic and Goidelic (from whence comes Irish)



Here's a useful document from Schrijver (it's on academia.edu but it's upside down)
http://compsoc.nuigalway.ie/~dubhthach/DNA/Pruners_and_trainers_of_the_Celtic_famil.pdf

dubhthach

There was some work done a number of years which put the spilt time between Brythonic and Goidelic at some stage around 800-1000BC. This would at least point to perhaps an origin in Atlantic Bronze age, with divergence between two perhaps been driven by the fact that Ireland appears to go into a "Dark age" for 500 years after 800BC or thereabouts.

markesmith

The Hekla 3 eruption happened about 1000 BC, and a severe temperature drop is recorded in the Irish pollen count.

I wonder if this caused a major depopulation in Ireland, making it that bit easier for the "Celtic" (for want of a better word) language to take hold?

Fascinating developments at the moment, and kudos for the concise explanations

dubhthach

markesmith wrote: »

The Hekla 3 eruption happened about 1000 BC, and a severe temperature drop is recorded in the Irish pollen count.

I wonder if this caused a major depopulation in Ireland, making it that bit easier for the "Celtic" (for want of a better word) language to take hold?

Fascinating developments at the moment, and kudos for the concise explanations

Well it's quite possible that dialects within the continum of Proto-Celtic were already in situ. Of course there are arguments that a non-Celtic language perhaps survived right up to 6th century. This is to do with argument over the word Partán(Crab) in Old Irish (modern Irish spelling is Portán) and the population group called the Partraige who were most concentrated in Connacht.

Proto-Celtic lacked a *p phoneme, as a result early borrowings into Irish from Latin for example underwent a p -> c shift (basically reflecting the sound change from /kw/ -> /p/ found in Brythonic and Gaulish).

So for example latin Pascha became old Irish Cásc. The fact that population had a "P" in their name had led people to conclude that they had been non-Celtic speaking (perhaps belonging to another branch of Proto-IE not affected by Celtic deletion of *p).

Leaving that aside if we go back to Schrijver's proposed divisions of Celtic languages into "North Celtic" and "South Celtic" another of things jump out.

"South Celtic" languages appear more "archaic", though obviously Lepontic underwent sound-change from /kw/ -> /p/ (Q-Celtic -> P-Celtic), probably due to influence of etruscan)

"North Celtic" includes Gaulish, Brythonic and Goidelic. The fact that Goidelic (from thence Irish, Scottish Gaidhlig and Manx descend) retained the ancestral /kw/ (Q) value, points to this sound-change been later. Perhaps in the Iron age.

Lepontic area came under Gaulish control in the Iron age so a possible difusion of sound change might be along the lines of:

Lepontic -> Gaulish -> Southern Britain

The fact that Ireland seems to be cut off for 500 years up until about 300BC might explain why the sound-change never made it across the Irish sea. In comparison there is documented evidence (Roman's etc.) of Gaulish input into Southern Britain in 2 centuries or so before Caesar ran amok. (That and tribes in Britain that minted coins are generally of more recent Gaulish origin).

Both Goidelic and Celtiberian remained outside of this sound-change zone, which basically reflects both areas status on periperhy when it came to trade networks etc of the Iron age if you ask me.