Review
of the Review of �The Land of Houlouf: Genesis of a Chadic Polity, 1900 BC �
AD 1800� .
Memoirs of the University of Michigan Museum of Anthropology No 38 by Susan Keech McIntosh (2005) in Antiquity 79 (304) June 2005
Introduction
There
has been at least four different reviews of this book, by Henry Tourneux in Mega-Tchad�
the Bulletin of the research network of the Chad basin), Scott MacEachern (American Antiquity 70: 194-5, 2005), G. Connah (African Archaeological Review 20: 171-4, 2003), and this one by Susan
McIntosh, whose web version has just been made available. The problem is not to
be praised or criticized. There are
important differences in emphasis between the four book-reviews alluded to above, but the
last one is special.
The
review by Susan Keech McIntosh stands out as the most abrasive and unfair for
unknown reasons that will be probed in this review of Reviews. The reader(s)
will allow the writer to use extensive quotes from the original review. But
before proceeding ahead, it is important to have a little background of the
situation. West Africa is huge; there is room for hundred of archaeologists
working side by side to explore the past of that part of the African continent.
Different ideas are explored and in fact researchers studying similar processes
would have been expected to learn more from each other. This writer has always
done and will continue to do that. For Susan Keech McIntosh on the other hand,
if judged from her published record, Holl�s West African research does not
exist, otherwise, it has to be erased.
The Review�s introduction
The
first segment the review introduction runs as follows: �Land of Houlouf reports primarily on findings from ten short field seasons
conducted between 1982 and 1991 at fourteen settlement mounds in Northern
Cameroon. These mounds are among the thousands distributed on the clay and sand
plains south of lake Chad in Nigeria, Cameroon and Chad. Relatively few have
been excavated in Nigeria by Connah in the 1960s and by an interdisciplinary
team from Frankfurt in the 1990s� .
After a quick allusion to the 4000 years long sequence and the
expression of high hopes, the book is summarized in a few lines: � The
monograph presents the research programme goals and data (pp. 27-202), set between
a brief background section and two concluding chapters that recast these data
into an evolutionary sequence that culminates with the emergence of a chiefdom
in the second Millennium AD�. That is precisely all that is provided to the
reader about the whole book.
A
textual analysis of the wording and structure of these opening statements
provides interesting insights:
(1)� findings from ten short field seasons�, (2) �most extensive and
reliable research�, (3) � a brief background section�, (4) �two concluding
chapters�. As can easily be
noticed, all these assertions are in the �diminutive� tone., but let us examine
them in more details.
(1)
It is not clear what is referred to by �ten short field seasons�. The writer�s
ten field seasons lasted for 4 to 8 weeks each; the 1989-90 field season was
split into two visits of 4 weeks each, in December 1989 and February-March
1990. All the other 9 field-seasons lasted for 8 weeks each, more or less the
standard length of annual field seasons. The choice of the word is very
interesting.
(2)
The meaning of � most extensive and reliable research� seems obvious uttered as
it is before reviewing the book. G. Connah excavated 7 sites, Bama Road Site, Bornu 38, Daima, Kursakata, Ajere,
Yaw, Birni Ghazzargamo, stretched
along a 1000 kilometer north-south transect. The Frankfurt-Maiduguri project
did the same, with excavations at Konduga, Gajiganna. Kursakata, Megge, Dikoa,
with however a more vigorous and coherent palaeoclimatological. Botanical, and
zoological components. None of these
project was framed as a tightly integrated regional project as the one
reported in The Land of Houlouf. All these are archaeological works but
the have different goals.
(3)
The expression �A brief background� lumps two chapters together to give the
impression that it is a shallow irrelevant part of the book. The book�s title
and structure are explained and made explicit in the preface on pages xv-xvi.
Chapter 1 (pp. 1 � 18) deals with the history of Chadian plain archaeology.
Chapter 2, (pp. 19-26) Social Formations..� delineate the theoretical
perspective adopted in the book. One of its aims is to move away from
unproductive terminological quibbling. And finally, chapter 3, (pp. 27 � 34)
spells out the research program, bringing chapter 1 and 2 into a pragmatic and
operational perspective.
Assertion
(4) is symmetric and complementary to assertion (3). The book surprisingly has
�two concluding chapters� because the reviewer is unwilling to call them what
they are: an integrated discussion of the emergence of complex social systems,
from seasonal camps to Kingdoms, framed in the Annales perspective
The
introductory statements presented above set the stage for the next step. For
the reviewer, �Land of Houlouf disappoints. It is painful to say this,
in view of the vast effort that went into the fieldwork and the preparation of
this volume�
On Excavation methodology
The
reviewer disappointment is then spelled out in the remaining part of the paper.
The second paragraph of the review identifies �several pervasive problems�. The
first to be mentioned revolves around the excavation methodology. The second is
that of �interpretive specificity�.
The allusion to faulty methodology is glossed upon, but never
elaborated. Inversion in radiocarbon dates are not a mystery; they can be
explained and are explained rationally wherever they occur. In fact they
provided very interesting insights into the �life� of the archaeological site
itself. Pits are dugs; other are filled. Erosion remobilized cultural remains
and dropped them somewhere else, etc� An archaeological site is not a
mausoleum. The misplaced dates are
not taken into consideration in the general reconstruction of the regional
sequence. The allusion to faulty methodology is expanded further, raising
doubts on the field competence of the writer. If the materials collected within
each 20 �90 cm deep occupation horizon were not �treated separately during
collection and reporting�, one can but wonder how the frequency distribution
and spatial scatter of shards as shown in the description of Houlouf Site would
have been possible. One would equally wonder how it was possible to separate
the material from habitation surface from that of the fill of collapsed and
reworked material. There are at least two tables per excavated site dealing
with a detailed taxonomy of the stratigraphic sequence. The sites sequence is
divided into �Sedimentary units� based on litho-facies. They are divided into
Occupation horizons that include a living surface, topped by a fill of
collapsed and reworked material.
Potsherds frequency, weight, density, and fragmentation indexes are used
to assess the formation process of the mound under consideration. A very
innovative work not only by West African standards. The issue raised at the end
of the paragraph, falls in the category of the disputed �interpretive
specificity�; the disagreement on the �horse house� suggestion according to
which �the very same deposits were interpreted in an earlier publication (Holl
1988: 25) as a secondary accumulation due to water erosion from an abandoned
sector of the site�. Unfortunately the reviewer reading of my previous
description is partial and inaccurate; it is severed from its context and
truncated. The idea developed in
the writer�s 1988 book was derived from the comparison of sherds fragmentation
indexes:
�Ainsi,
les tessons des unites stratigraphiques comprises entre 1,60 et 2,00 m
rsulteraient de laccumulation provoqu�e par le ruissellement dans un secteur
abandonn� du site. Les indices de fragmentation compris entre 10,36 et 10,41
semblent indiquer un pi�tinnement de faible intensit� (Holl 1988: 25)
Translated
as:[.. Accordingly, shards from stratigraphic units exposed between 1.60 and
2.00 m may have been accumulated by water run-off from an abandoned part of the site. Fragmentation indexes,
varying from 10.36 to 10.41 tend to indicate a low degree of trampling].
In
addition, any archaeological feature takes its potential functional
interpretation from its context and the cultural remains it is associated with.
The idea of 钓horse hut� is inferred from the watering-trough, the dung shape
(round hut), and evidence of horsemanship found with the adult skeleton buried
in what was part of the courtyard, with dispersed complete vessels. The
inference is not as ad-hoc as the reviewer intends it to be understood.
No Data in Charts and Tables?
The
third paragraph starts with a wish, that �the excavation data from the fourteen
sites were clearer and more user friendly�. But as can be expected the reviewer is disappointed once
again: �Instead of presenting quantitative and descriptive data on artifacts
and fauna in charts and tables, the author embeds them in a dense narrative,
that consider each occupation horizon and excavation unit in turn� . It was
precisely the intention of the writer to discuss archaeological evidence within
its context of finds before moving toward higher generalization. However, the
next assertion according to which �For the numerous urn burials excavated,
there is no summary table of grave goods, urn types, body position, age and
sex, ..�, etc.. flies in the face of the evidence. A listing all the
grave-goods from the urn burials is found in table 37 (p. 218). Details from
the pottery is found in table A7, to which the reader is referred to several
times on pages 176-177. The
position of the body is identical for all the burials from the Houlouf phase cemetery. �The 24 deceased uncovered
in the cemetery were buried in almost the same position: facing southwest, in a
sitting or almost upright
position(..), and 21 with their feet in pots� (Holl 2002: 177). All the
deceased are adults and the state of preservation of bones did not allow for
the determination of sex. There is
nothing wrong with a reader having to search for the information he/she wants.
That is what library research is all about. It is surprising to required from
any potential writer to provide �ready to cook� data files.
The
reviewer is well known for her advocacy of pottery sequences and it is from
this background that one can understand her comments on shards �not studied at
all�. First of all, sherds were used in the study, not build a pottery sequence
but to assess site formation processes. Second, with few exceptions, there are
enough complete vessels or restorable size pieces that are used to bring to
light the potting traditions that developed throughout the 4000 years long
archaeological sequence. The aims of the research is to reconstruct the
patterns of craft practices that developed in the past and follow their change
/or lack of it through time. The issue is not to stick to a specific method.
It
can easily be seen that most of the assertions uttered in paragraph 3 are
incorrect.
A dubious Chronological Framework
The
reviewer has her strongest words in the four paragraph of her review. For the
use of radiocarbon dates what matters really is not their quantity but their
distribution, in combination with stratigraphic information. It is as simple as
that. The phasing of the regional sequence is �based on major changes in
settlement pattern. Each settlement was named after the site from which the
most significant archaeological record from the phase has been recorded� (Holl
2002: 203). Not surprisingly, the
conclusion is inescapable. � In the end, one must conclude that the ambitious
theoretical agenda persued is not sustained by the data presented��
Conclusion:
The
writer does not wish to second guess the intentions of Dr. Susan Keech
McIntosh. All along this review of the review, it has been shown quite
systematically that her assertions are incorrect. Her conclusion is even more
troublesome if one reads the review by G. Connah, the research she cites as
exemplar. He has his own critical stance on The Land of Houlouf
and his development are particularly interesting:
�These
criticisms must be balanced against the undoubted merits of this book. For
instance, the recognition of Occupation Horizons in the excavations and their
detailed description is very important and goes far beyond the mechanical recording
of arbitrary spits or the identification of more general units of natural
stratification that have often been the case on such sites. In particular, the
analysis of those surfaces, with their remains of dwellings and evidence of
domestic, industrial, and ritual activities, and the application to them of
radiocarbon dating, allows the construction of a series of sites sequences that
relate to actual occupations and form components in an overall sequence for the
area. In the end the book really does live up to its title, managing to trace
the process by which intermittent seasonal sites gradually evolved first into
autonomous villages, then into a cluster of villages with two dominant centers,
and later into a chiefdom where one large city dominated the rest of the
settlements, only to be integrated finally into an even larger state(..) Also
deserving credit is the ambitious analysis of the Houlouf phase A cemetery, of
AD 1500-1600 (�), particularly for its attempt to assign rank and occupation to
the different burials� (Connah
2003: 173).
The
Land of Houlouf is a highly
innovative work; it generates its
own problems but it stands out as a unique case of strongly integrated regional
analysis and empirically sound study of a specific case of the emergence of
complex social systems. Dr. Susan Keech McIntosh has a different take. So be
it.
Augustin
F.C. Holl
Museum
of Anthropology
The
University of Michigan,
Ann
Arbor
�