A PRELIMINARY INVESTIGATION OF THE BLOOD GROUPS OF THE SAB BONDSMEN OF NORTHERN SOMALILAND DR. K. L. G. GOLDSMITH, Westminster Hospital

A PRELIMINARY INVESTIGATION OF THE BLOOD GROUPS OF THE SAB BONDSMEN OF NORTHERN
SOMALILAND
DR. K. L. G. GOLDSMITH, Westminster Hospital
DR. I. M. LEWIS, University College of Rhodesia and Nyasaland



While the physical characteristics of free-born noble Somali have been studied,
as far as is known no attention has been given to the sab bondsmen.(1) This
paper reports the results of a preliminary serological investigation of a small
sample of 54 male sab. and compares the results with those of a study of 1000
noble Somali.(2) Samples of blood were obtained with the co-operation of the
Department of Medical Services of the Somaliland Protectorate,(3) and with that
of Midgaan and Tumaal elders of Hargeisa town. Our results treat these two
groups as one sample.
The Midgaan (pl. Midgo) (hunters, leather-workers, professional barbers, etc.),
Tumaal (pl. Tumaalo) (mainly blacksmiths), and Yibir (pl. Yibro) (itinerant
pedlars and magicians) known to free-born Somali as sab. (4) have frequently
been described as 'low caste' but they are more correctly bondsmen.(5) The
distinction is usually phrased as between sab iyo Soomaali (sab. and Somali);
or Aji iyo Midgo (noble Somali and Midgaan). Traditionally each member of this
class was bound in servitude to a noble Somali family. Not every noble Somali
had a sab. bondsman attached to him but all sab were subject to a noble patron
and protector. To a large extent this pattern persists in Northern Somaliland,
although many of the sab. have achieved a partial emancipation by moving to
practise their special skills in the towns and urban centres. Out of a total
Somali population of about 2,500,000 in the British Protectorate and Somalia,
the total number of sab bondsmen is probably at the present time not more than
12,500. The Midgaan (9,000) are the most numerous, with the Tumaal (2,250) next
and the Yibir (1,300), Who are now rare, last.
The sab. live scattered all over Northern Somaliland in real or putative
patrilineages, on a pattern similar, though reduced, to those of their noble
Somali (Aji) protectors. In different regions they are known locally by the
names of the eponyms of the small agnatic lineage groups into which they are
divided, rather than by their occupational classification as Midgaan, Tumaal
and Yibir. Ultimately, however, it is as these and, collectively, as sab, that
they are described. In the north the two largest Midgaan lineage groups are the
Muuse Diiriye in the west, and the Madiban in the east. Except through their
protectors, sab. have traditionally no rights in the field of Somali political
relations. Nowadays a few sab. who have succeeded in amassing considerable
wealth as traders, or who have enhanced their status by employment in
government service, have achieved the right to attend and speak at the councils
of the Somali lineage groups to which they are attached. But noble Somali still
do not marry with them. Illicit unions sometimes occur, and a noble Somali may
beget a child by an attractive sab woman. The Midgaan, Tumaal and Yibir marry
amongst themselves but not with Aji Somali. Most of their activities are
amongst themselves, and they have relations with noble Somali only indirectly
through their patrons. They have their sheikhs and wadaad (6) who fulfill for
them the same functions as do those of their masters amongst noble Somali. All
sab speak the dialect of the Somali to whom they are attached. But they have
also secret codes, hardly sufficiently extensive to be called languages, (7)
which they use when they wish to conceal what they are saying from listening
Somali. The sab themselves, and the Somali - who would not bemoan themselves by
troubling to learn them - do not regard these dialects very seriously.
Apart from their specialist trade skills their cultural features are identical
with those of Somali. The Midgaan it is true, have songs of magical intent
associated with hunting animals, but this is part of their specialist
knowledge. Some of the sab, particularly the Yibir, who have a traditional
right to collect alms from Somali, used to be called derogatively
'corpse-eaters' (bakhti cune) because they were alleged to eat meat considered
impure by Somali. But there is little indication that this is still true today.
Various traditions are current in Somaliland of the origin of the sab.
According to some Somali, the sab descend from noble Somali who became degraded
by eating meat which had not been properly slaughtered in Muslim fashion during
a famine. Some sab informants also referring to well documented cases, maintain
that they are of the same stock as the Somali but descend from small,
numerically weak lineage groups which were reduced to servitude by more
powerful enemies. (9)
In support of these claims, Midgaan informants have produced genealogies
tracing descent from Dir, the founder of the Dir clan family, generally
regarded as the oldest Somali stock. Some Tumaal trace descent from Daarood
founder of the noble Daarood Somali clan family. Various writers have suggested
that the sab represent the remnants of pre-Somali peoples conquered by Hamitic
Somali invaders.(11) In Dr. Cerulli's view, however, whatever common
characteristics the sab possess are to be ascribed not to common ethnic origin,
but to the action of common historical processes. (l2) Cerulli regards them as
a mixed conglomerate people of partly pre-Hamitic origin.
Since there are no main cultural differences, and no strong traditions of
diverse origin, (13) it would seem that only comparative study of their
physical characteristics can throw light on their origins. Miscegenation is
almost if not entirely, forbidden, so that were the sab originally possessed of
different physical characters to the Somali these differences should still be
evident. To the casual observer most sab in Northern Somaliland look much the
same physically as Somali. It is sometimes possible, however, to identify a
Midgaan, Tumaal or Yibir. Somali indeed usually claim to be able to distinguish
sab, but it is difficult to decide whether or not they do this only on a basis
of physical features. In the presence of Somali, sab tend to adopt a
subordinate bearing and presence. As was stated at the outset no physical study
of the sab has yet been made as far as we know. The results of our serological
investigation are set out in the following tables. They show that there is no
significant difference in the distribution of the ABO and MN groups in our
sample and in a sample of 1,000 noble Somali. The contingency table for the
Rhesus group does not accord so exactly, but the difference is small and may be
due to the size of our sample. As far as the results go, the blood group
composition of our sab sample is thus virtually identical with that of the
Somali. While it would be dangerous to generalize from so small a sample (54
individuals), our results suggest that in serological characters there is no
difference between those of the sab and those of noble Somali. Before a
definite conclusion can be reached, however, further and larger samples of sab
blood are required. It would be valuable to have samples from different parts
of Somaliland, for it cannot be assumed that sab from other parts of Somaliland
have the same serological characteristics as our Hargeisa sample.
Table I. THE "ABO" GROUPS OF TUMAAL AND
MIDGAANGroupOA1A2BA1BA2BTotalOBTAINEDNumber26124110154.00Per Cent
48.1522.227.4120.370.001.85100.00GENESrp1p2qTotal
0.70500.11810.05690.12001.0000
Table II. CONTINGENCY TABLE TO COMPARE "ABO" BLOOD GROUPS OF TUMAAL AND MIDGAN
ON THE ONE HAND WITH 1,000 SOMALIS ON THE OTHERBlood GroupMidgaan and
Tumaal1000 Somalis
TotalObtainedExpectedObtainedExpectedO2630.74574569.26600A1127.28130134.72
142B119.07166167.93177A2+A1B+A2B56.91130128.09135Total5454.001,0001000.001,¬054x
to the power 2=4.9853. For a value of x to the power 2=4.9853 where n=3, the
probability P lies between 0.20 and .10.
Table III. The "MN" GROUPS OF TUMAAL AND MIDGAAN GroupMMMNNNTotalOBTAINED
Number16231554.00Per Cent29.6342.5927.78100.00GENESMNTotal0.50930.49071.0000
Table IV. Contingency Table to Compare Rhesus Blood Groups of Tumaal and
Midgaan on the One Hand With I,000 Somalis on the OtherRH GroupsMidgaan and
Tumaal1000 Somalis TotalObtainedExpectedObtained
ExpectedMM1613.68251253.33267MN2327.31510505.69533NN1513.01239240.98254Tot
al5454.001000.001,054x to the power 2=1.4524. For a value of x to the power
2=1.4524, when n=2, the probability P lies between 0.50 and .300.
TABLE V. THE RHESUS-GROUP RESULTS OF TUMAAL AND
MIDGAANGroupCCDeeCcDEeccDEECcDeeccDEeccDeeccDueeccddeeTotalObtainedNumber0
32003233254Percent0.005.5637.030.005.5642.595.563.70100.00ChromosomesCDecD
E>cDuecdeTotal0.21300.05560.11030.19541.000GENESCE0.21300.0556
TABLE V1. CONTINGENCY TABLE TO COMPARE RHESUS BLOOD GROUPS OF TUMAAL AND
MIDGAAN ON THE ONE HAND WITH I,000 SOMALIS ON THE OTHERTypeMidgaan and
Tumaal1000 Somalis TotalObtainedExpectedObtainedExpectedCCee &
Ccee2014.35260265.65280ccEE &
ccEe612.60240233.40246ccee2827.05500500.95528Total5454.001,000.0010001,054x to
the power 2=6.0237. For a value of x to the power 2=6.0237, when n=2, the
probability P lies between 0.05 and 0 .02.>
Notes
1.Eg. P. Lester, 'Etude anthropologique des populations de l'Ethiopie,'
L'Anthropologie, Vol. XXXVIII (I928), p. 289; id., 'Contribution a
l'anthropologie des Somalis, ' Bull. Me'nt. Soc. Anthrop. de Paris, Vol. VIII
(1927), p. I75; D. R. Parenti, 'Antropologia della Somalia Meridionale,'
A.A.E., Vols. ...