SOMERLED, Lord of the Isles

[user]

This is his pedigree as I've found:

Gille Adomnan
Gillebridge
Somerled (1030-1083)
Gille Adomnan
Gillebridge
Somerled II, Lord of the Isles (-1164)

This repetition of three names looks very suspicious and I wonder if the
author goofed. Can anyone confirm or deny this?

[Stewart Baldwin]

user <ste...@pss.fit.edu> wrote:

You are right. They are suspicious. In February 1996, I posted an
item to s.g.m which touched on a couple of topics, one of which was
the ancestry of Somerled. The part of that item which is relevant to
Somerled is quoted below:

---------------

The best account of the ancestry of Somerled of which I am aware
is the article "The origins and ancestry of Somerled", by W. D. H.
Sellar, which appeared in "The Scottish Historical Review", vol. 45
(1966), pp. 123-142. This article lists all of the earliest known
manuscripts giving the genealogy of Somerled, and then discusses the
reliability of the genealogy in detail. To make a long story short,
he accepted the genealogy as being probably reliable back to the
early ninth century, as follows:

1. Fergus, probably a member of the minor Irish sept of Ui Maic Uais.

In his article, Sellar rejected as "preposterous fiction" the claim
that Fergus was the same person as the king of Dalriada of that name
who reigned from 778 until 781. His exact line of descent from the
Ui Maic Uais is unknown. (The traditional genealogy of Fergus is
chronologically impossible by several hundred years.)

2. Gofraidh (Godfrey), son of Fergus, whose name betrays a likely
Norse connection, said by the Annals of the Four Masters to have gone
from Oirgiall to Alba (Scotland) at the request of Kenneth MacAlpin
in 836. Godfrey's death is given by the Annals of the Four Masters
as 853. The Annals of the Four Masters are a very late compilation,
and not always to be trusted, but Sellar argued that they should be
accepted as reliable in this case.

3. Maine. (Nothing is known of the next several generations other
than their names.)

4. Niallgus.

5. Suibne.

6. Meargaige. Some have attempted to identify this person with king
Eachmarcach of Dublin, but Sellar shows that this is incorrect.

7. Solam (Solomon).

8. Gilla-Adomnain.

9. Gilla-Brighte.

10. Somerled, d. 1164.

Stewart Baldwin

[Rudy Krutar]

user wrote:
>
> This is his pedigree as I've found:
>
> Gille Adomnan
> Gillebridge
> Somerled (1030-1083)
> Gille Adomnan
> Gillebridge
> Somerled II, Lord of the Isles (-1164)
>
> This repetition of three names looks very suspicious and I wonder if the
> author goofed. Can anyone confirm or deny this?

User,

What was your source? Was it better than mine?
We might work out what did happen by tracing sources.
Those who urge you not to cite imperfect sources are absurd.

As you can see below, Somerled's ancestry is far from certain.
Solmund could be confused with Somerled. And his father's name,
could easily be confused: was it Echmarcach, Imergi, Meargach,
or Ineargach? These could be variations of the same name in
different languages or dialects or they could be names of different
people.

Rudy Krutar, descendant of Somerled

I have the following on the ancestors of Somerled:
_____________________________________________
Ancestors according to Norse tradition as sketched in an undated
manuscript titled "Clan Donald" with no author listed are marked
with Å's. Ancestors according to Irish tradition are marked with A's
whenever the traditions differ.

(The Ahnenriehe numbers below are in hexadecimal
for easier arithmetic -- twice 8 is 10).

Å+20000. SOMERLED MacGillebride, eighth Thane/Regulus/Lord
of Argyll, lord of Cantyre, lord of the Hebrides, founder of the
"Kingdom of the Isles", Norse King of the Sudreys; first completely
historical chief of what would become Clan Donald; # 1164; * Abbey
of Saddell in Kintyre; son of:
Å+40000. Gillebride / Giolla Brighid, chief of the Royal House
of Argyll; # 1164, Renfrew SCT.
Å+80000. Gilladamnan / Giolla Adhamnan; married into the
Royal House of Argyll.
Å+100000. Solmund / Solamh(Solomon); married into family
of St. Columba, whose emblem was a hand holding a cross.
Å+200000. Echmarcach/Imergi / Meargach/Ineargach, King of Dublin;
# 1065 on pilgrimage to Rome

* O'Hart, Irish Pedigrees, vol.I, p.528.
* "Clan Donald", p. 4.

Å+200000. ECHMARCACH/Imergi / MEARGACH/Ineargach,
King of Dublin; defeated in Man in 1061;
# 1065 on pilgrimage to Rome; his ancestors
(10% sure as based on similarity of names and arms)
according to "Clan Donald":
Å+400000. Ranald, King of Waterford, 1022-1031;
# 1035, killed in Dublin.
Å+800000. Ranald; # 995.
Å+1000000. Ivarr, King of Waterford and Dublin; # 1000.
Å+2000000. Ivarr; # 950.
Å+4000000. Ranald, King of Dublin, Waterford, and York; # 921.
Å+8000000. Guthorm, King of Dublin?.
Å+10000000. Ivarr, King of Dublin; sacked Dunbarton in 870.
Å+20000000. King Godfrey; raided Ireland in 854; # 873.
Å+40000000. King Ranald "Higher-than-the-Hills";
expelled to Orkney.
Å+80000000. Olaf "Geirstada-Alf", King in Vestfold, 810-840.
Å+100000000. Godfrey "the Proud", King in Vestfold,
Raumarike, and Vestmarar; # 810.
Å+200000000. Halfdan "the Stingy", King in Vestfold, 8th century.
Å+400000000. Eystein "the Fart", King in Raumarike, 8th century.
Å+800000000. Halfdan "White-Leg", King in Vermaland in Norway.
Å+1000000000. Olaf "Tree-Hewer"; # 710 ca, sacrificed.
Å+2000000000. Ingiald "Ill-Ruler", last of the pagan sacral
peace kings of Uppsala in Sweden

* "Clan Donald", pp. 3-4.

(The %-ages below provide uncertainty factors for fuzzy logic,
not probabilities for Bayesian analysis -- we cannot repeat experiments
here.)

MEARGACH's ancestors were Irish (20% sure as the MacDonalds
considered themselves as the Race of Conn, which is a patrilineal
heritage in Gaelic tradition, but 10% sure as too few generations
are shown -- one per century) according to O'Hart:
A+400000. Suibhneach.
A+800000. Niallgus.
A+1000000. Main.
A+2000000. Gothfrith.
A+4000000. Fergus (same as A+40051040000.?).
A+8000000. Earc, (5th century?).
A+10000000. Carthann (charity, friendship, kindness);
ancestor of many saints.
A+20000000. Earc/Eachach.
A+40000000. Eochaidh.
A+80000000. Colla Uais, 121st Monarch of Ireland; •••.

* O'Hart, Irish Pedigrees, vol.I, p.527-528 "MacDonnell".

[D. Spencer Hines]

Stewart Baldwin wrote:

<snip>

> There is a another gap here. The parentage of Ivar of Waterford is
> unknown. It was once suggested by J. H. Todd (editor of "Coghad
> Gaedhel re Gallaibh", London, Rolls Series, 1867, p. 294) that Ivar of
> Waterford was a grandson of Ivar (d. 950, Annals of the Four Masters),
> and Todd's genealogical table has a dotted line between the two Ivars
> to indicate this suggested descent (which was no more than a
> conjecture on Todd's part, as he clearly stated). It appears that
> someone has converted the dotted line of Todd's table into a solid
> line without even bothering to read Todd's comments.
> =

> > =C5+2000000. Ivarr; # 950.
> =

> Another gap. Same reason. Todd's table has a dotted line, indicating
> that he thought Ivar might have been a descendant of Ragnall (in some
> way unspecified by Todd), and this conjecture has been carelessly
> converted into a supposed father-son relationship.

<snip>
> =

> >MEARGACH's ancestors were Irish (20% sure as the MacDonalds
> >considered themselves as the Race of Conn, which is a patrilineal
> >heritage in Gaelic tradition, but 10% sure as too few generations
> >are shown -- one per century) according to O'Hart:

> =

> I'm puzzled by your percentages. Shouldn't "completely impossible"
> (which one generation per century certainly is) count as 0% sure (or
> 100% sure of falseness)?
> =

<snip>

> This is an extremely poor source for Irish genealogies, and all
> statements made by O'Hart should be taken with a grain of salt.
> =

> Stewart Baldwin

As the sorry examples above indicate, much of what passes for
authenticated---very early---Medieval Genealogy should be taken with a
grain of salt. People are still converting "dotted lines" into "solid
lines" to this day-----including suposedly reputable professional
Genealogists.

By the same token, there is absolutely no doubt that supposedly
intelligent researchers leap to conclusions without reading and
cogitating over the appended comments----in Kay Allen's funny, but
colorful, malapropism --- these are "conclusion longjumps"---and the
peculiar world of Genealogy is heavily populated with them----because
people just, "So Want To Believe That They Are Descended From
Royalty"---and the facts and evidence be damned.

Caveat Emptor.

-- =

D. Spencer Hines----"City life breeds a mass mind 'hostile to reason,
shifty, irritable, credulous and violent', easy prey for conspiracy
theories and delusive promises of racist, imperialist or socialist
redemption." Robert Skidelsky, "The Road From Serfdom", Viking Penguin
(1996), p. 39

[Stewart Baldwin]

Rudy Krutar <kru...@his.com> wrote:

>User,

In a separate posting, I have already outlined the ancestry of
Somerled as given by W. D. H. Sellar in his article "The origins and
ancestry of Somerled" [The Scottish Historical Review 45 (1966),
123-142], which is essentially the same as the second option listed
here down to Goffraid mac Fearghus (with a few spelling variations on
the names). Below are a few comments on the other data given here.

>(The Ahnenriehe numbers below are in hexadecimal
>for easier arithmetic -- twice 8 is 10).

>Å+20000. SOMERLED MacGillebride, eighth Thane/Regulus/Lord
>of Argyll, lord of Cantyre, lord of the Hebrides, founder of the
>"Kingdom of the Isles", Norse King of the Sudreys; first completely
>historical chief of what would become Clan Donald; # 1164; * Abbey
>of Saddell in Kintyre; son of:

There is no evidence to justify calling Somerled "Norse".

> Å+40000. Gillebride / Giolla Brighid, chief of the Royal House
> of Argyll; # 1164, Renfrew SCT.

The death date of 1164 is doubtful, as is the claim that he was
"chief" of the royal house of Argyll..

> Å+80000. Gilladamnan / Giolla Adhamnan; married into the
> Royal House of Argyll.

In would be interesting to see a good source for this alleged
marriage.

> Å+100000. Solmund / Solamh(Solomon); married into family
> of St. Columba, whose emblem was a hand holding a cross.

Ditto.

> Å+200000. Echmarcach/Imergi / Meargach/Ineargach, King of Dublin;
> # 1065 on pilgrimage to Rome

This identification of the Meargaige (father of Solam) of the
pedigrees of Somerled with king Eachmarcach of Dublin is incorrect.
It is an unsupported guess which was apparently based on the false
assumption that the king Margad of Dublin who appears in some Norse
sources was Eachmarcach, whereas it is now known that "Margad" was a
Norse from of the Irish name "Murchad", and that there was a king
Murchad of Dublin in the Irish sources who is probably the Margad of
the Norse sources. Thus, Margad of Dublin was not Eachmarcach, and
the similarity of the name Meargaige (and variants) can not be used to
support the guess (and it was never anything more) that Meargaige and
Eachmarcach were the same person. Since Meargaige and Eachmarcach
were different individuals, none of the rest of this part of the first
genealogy is relevant to the ancestry of Somerled.

>* O'Hart, Irish Pedigrees, vol.I, p.528.
>* "Clan Donald", p. 4.

>Å+200000. ECHMARCACH/Imergi / MEARGACH/Ineargach,
>King of Dublin; defeated in Man in 1061;
># 1065 on pilgrimage to Rome; his ancestors
>(10% sure as based on similarity of names and arms)
>according to "Clan Donald":

There is a gap here. All that is known is that Eachmarcach's father
was named Ragnall, but it is not known which of the several Ragnall's
who appear in the records of this period was the father of
Eachmarcach. I posted an item in s.g.m awhile back called "Too many
Ragnalls" which discussed this in some detail.

> Å+400000. Ranald, King of Waterford, 1022-1031;
> # 1035, killed in Dublin.
> Å+800000. Ranald; # 995.
> Å+1000000. Ivarr, King of Waterford and Dublin; # 1000.

There is a another gap here. The parentage of Ivar of Waterford is

unknown. It was once suggested by J. H. Todd (editor of "Coghad
Gaedhel re Gallaibh", London, Rolls Series, 1867, p. 294) that Ivar of
Waterford was a grandson of Ivar (d. 950, Annals of the Four Masters),
and Todd's genealogical table has a dotted line between the two Ivars
to indicate this suggested descent (which was no more than a
conjecture on Todd's part, as he clearly stated). It appears that
someone has converted the dotted line of Todd's table into a solid
line without even bothering to read Todd's comments.

> Å+2000000. Ivarr; # 950.

Another gap. Same reason. Todd's table has a dotted line, indicating
that he thought Ivar might have been a descendant of Ragnall (in some
way unspecified by Todd), and this conjecture has been carelessly
converted into a supposed father-son relationship.

> Å+4000000. Ranald, King of Dublin, Waterford, and York; # 921.

Ragnall (d. 921) was the grandson of Ivar, but there is no solid proof
for the identity of the intervening generation.

> Å+8000000. Guthorm, King of Dublin?.

I know of no king of Dublin during this period who had this name (or
any similar name). There was a king Guthrum of the Danes of East
Anglia and a king Guthred (or Guthfrith) of York (said to be father of
Ragnall by Adam of Bremen, whose source and reliability are uncertain
on this matter), but neither was king of Dublin. I imagine that it is
the latter who was intended here.

> Å+10000000. Ivarr, King of Dublin; sacked Dunbarton in 870.

He died in 873. You can forget the earlier "generations" of this
pedigree. It looks like someone has taken the pedigree of Ivar which
appeared in the Fragmentary Annals of Ireland (often called "Three
Fragments") and tortured it to fit part of the standard Heimskringla
pedigree. Pure fantasy.

> Å+20000000. King Godfrey; raided Ireland in 854; # 873.

If he existed at all (and he appears only in the Fragmentary Annals),
there is no reason to believe that he was the son of the Rognvald in
whose honor Ynglingatal was composed.

> Å+40000000. King Ranald "Higher-than-the-Hills";
> expelled to Orkney.
> Å+80000000. Olaf "Geirstada-Alf", King in Vestfold, 810-840.

etc. [several names from Heimskringla snipped]

Judging from the above, I am not inclined to place much trust in this
obscure "Clan Donald" manuscript which you are using as a source for
this.

>(The %-ages below provide uncertainty factors for fuzzy logic,
>not probabilities for Bayesian analysis -- we cannot repeat experiments
>here.)

>MEARGACH's ancestors were Irish (20% sure as the MacDonalds

>considered themselves as the Race of Conn, which is a patrilineal
>heritage in Gaelic tradition, but 10% sure as too few generations
>are shown -- one per century) according to O'Hart:

I'm puzzled by your percentages. Shouldn't "completely impossible"

(which one generation per century certainly is) count as 0% sure (or
100% sure of falseness)?

> A+400000. Suibhneach.

> A+800000. Niallgus.
> A+1000000. Main.
> A+2000000. Gothfrith.
> A+4000000. Fergus (same as A+40051040000.?).

Gofraidh mac Fearghus appears in the Annals of the Four Masters under
the years 836 and 853. Sellar considered the genealogy to be probably
correct back to this point. The supposed link to the earlier
generations is obviously chronologically impossible. As discussed by
Sellar, the early genealogies of the sept of Ui Maic Uais (from which
Gofraidh mac Fearghus seems to have descended) are in a great state of
confusion. The following individuals, if they existed at all, lie
before the historically documented period.

> A+8000000. Earc, (5th century?).
> A+10000000. Carthann (charity, friendship, kindness);
> ancestor of many saints.
> A+20000000. Earc/Eachach.
> A+40000000. Eochaidh.
> A+80000000. Colla Uais, 121st Monarch of Ireland; •••.

>* O'Hart, Irish Pedigrees, vol.I, p.527-528 "MacDonnell".

This is an extremely poor source for Irish genealogies, and all

statements made by O'Hart should be taken with a grain of salt.

Stewart Baldwin