British

"The Harris Matrix" and "single-context recording" are now fundamental tools of archaeological excavation. But where did it all start? British Archaeology asked Edward Cecil Harris to look back on what he calls his "outstanding British invention".

The matrix was invented in Winchester, in the late evening of February 28 1973. It was an artistic, or visual epiphany, which signalled a major paradigm shift in archaeology. It might be more proper to claim this British invention as a mid-Atlantic revolution, as I was born, raised and mostly reside in Bermuda, of Bermudian-American parentage.

I raise that point of ancestry, as it is perhaps relevant to the acceptance or rejection of the matrix and its allied concepts around the world. In the United States, for example, the matrix has made slow in roads, probably due to cultural prejudice. Conversely in Britain, some view me as an American. The first major conference on the matrix was held in September, not in London, but at the University of Vienna, due to the advocacy of Wolfgang Neubauer.

In the early days Antiquity refused to publish an article on the matrix. American Antiquity also refused to publish one of my papers, and one reviewer was so incensed by it that he or she advised, in effect, that I be booted out of archaeology. As far as I know, nothing of substance on the matrix and its fundamental concepts has yet to appear in either journal. It was not until 2003 that a Harris Matrix stratigraphic sequence diagram appeared in the premier Antiquaries Journal, related to research at Swalecliffe in Kent by Robert Masefield and others. Many prehistorians, in particular, yet adhere to the notion that if stratigraphy is present on "their" sites, it is only of geological import.

It has always been my view that the Harris Matrix could equally be applied to archaeological and geological stratification. I recently received an email from geologists at the University of Liverpool who had developed what they call "younging tables" to display stratigraphic sequences, apparently meeting with considerable "customer resistance". I should advise these allied stratigraphic scholars that it is likely that they will have to wait 30 years for wider acceptance of their proposed paradigm shift. That is the generation gap, or die-off, that is sometimes needed for change to take place, as suggested by Thomas Kuhn, in "The Structure of Scientific Revolutions" in 1962. So what, in the face of apparent indifference in some quarters, makes the matrix so significant?

Once I had drawn up a group of little rectangular boxes, while doodling 35 years ago, I realised that I had an image into which stratigraphic data could be inserted, and that the whole world of archaeology had irrevocably changed. I named the image, somewhat in jest, the "Harris Matrix". A couple of days later I discussed its potential with Martin Biddle, an excellent field worker and stratigrapher who had founded the Winchester Research Unit in 1961, and from whom I learned much about excavation.

The revolution value of the matrix became apparent when it was understood that it produced the true "stratigraphic sequences" of archaeological sites. The diagrams were in effect four-dimensional, each numbered box representing the three dimensions of a deposit or surface, and containing the time components of each such stratigraphic unit. Lines representing the relationships of these units in relative time connected all the boxes.

The Harris Matrix changed the paradigm of archaeology from the two dimensions of the section as the profession's stratigraphic icon, to the four dimensions represented in the unique sequence of every site. Like the clock face and the Gregorian calendar, the matrix allowed everyone, as never before, to see time on archaeological sites, in diagrams against which all post-excavation analyses and reconstructions should be compared.

The point is that I knew from its invention that the Harris Matrix was "true", and that if the diagram produced in it was rubbish, it was a case of the old computer adage of "garbage in, garbage out" – a fine pun formuch stratigraphic work in archaeology even today. But what were archaeologists not doing correctly in stratigraphic recording, that resulted in so much trash being produced out of the scientific excavation and destruction of sites? It took five years to find out.

Four names stand out. Frances Lynch kindly corresponded with me regarding the crucial nature of what I termed "interfaces". Lawrence Keen, who pushed for what was eventually called "single-layer planning", a major facet of the interface concept, deserves thanks and credit for that seminal idea. Finally, David Wilson and James Graham-Campbell, my PhD tutors at University College London, allowed me to write my thesis on the principles of archaeological stratigraphy.

In 1973, I called the original matrix diagrams "layer charts", an indication of the continuing obsession with the material deposit as the central stratigraphic unit; I had not then appreciated the importance of the immaterial surface, or interface. Archaeologists had long over-looked the importance of the surface in stratification. Surfaces are the means by which stratigraphic sequences are composed, the implication outlined by Sir Mortimer Wheeler and Dame Kathleen Kenyon in the post-war controversy over whether section drawings should record surfaces by hard or soft lines. Deposits support surfaces in such analysis, but it is only through the study and full record of surfaces that a stratigraphic sequence can be compiled. It was not however until Keen that there was an insistence that all surfaces had to be recorded. I believe the first site to be so treated was excavated by Patrick Ottoway at Winchester, atmy suggestion in 1975.

Surfaces will always dominate: at any archaeological site, there will be surfaces with associated unexcavated deposits, or others (such as interfaces) that have no associated deposits. So if surfaces were not, or are not individually recorded,most of a site's stratigraphic units are lost.

In the science of archaeology, after destruction by excavation it should be possible to reconstruct a site through the record, the archive on paper or computer. The fundamental building blocks of such reconstruction are the records of the surfaces. Without them it is impossible, as a site's topographical nature forms the essence of a "phase" or "period" in its development. No amount of section analysis, or "piece plotting" of artefacts can reconstruct the Humpty Dumpty of the site. It is the surface of the egg, and its record, that hold the keys to reconstruction, with secondary contributions for absolute time coming fromdeposits – the yoke and egg white, if you will. Neubauer has also suggested that the "bottom" surface of each deposit must be recorded, to give one the complete 3D envelope of the deposit and its entire surface.

Surfaces can be recorded in their entirety, whereas deposits can only be sampled: they are mostly by necessity thrown away. With GIS and computer mapping, there can be no excuse for not recording all surfaces on a site, to provide the "layers" for its topographical reconstruction. The matrix tests and controls the analysis of the artefacts, and the ultimate position of surfaces in the "phasing" of sites, with multi-linear stratigraphic sequences. With no records of surfaces to give faces to phases, the reconstruction of an archaeological site is unobtainable.

I have given some space to surfaces to emphasise their critical value, but also to indicate that the matrix cannot work without other concepts: it was but part of the intellectual construct that I brought together. Single-context planning and the Harris Matrix are the horse and carriage of stratigraphic recording, with the artefacts under lock and key in the boot.

The matrix was invented at Winchester, but it was developed in London. I remain grateful to Brian Hobley and the teamfromthe former Department of Urban Archaeology who were willing to experiment with the concepts on the GPO London site in 1974 onwards. While I researched archaeological stratigraphy at UCL, they proved the efficacy of the system.

Principles of Archaeological Stratigraphy, the outcome of my research, appeared as a textbook in 1979. The publishers would not issue a paperback edition, and the hardback remained uncommonly expensive. To make it more widely available, early in the new millennium I bought out the remaining copies, acquiring the copyright. Thanks to Neubauer and Klaus Loecker of the University of Vienna, since April the book has been available digitally (see end note). Already some 3,000 free downloads have been made fromover 90 countries. Many have emailed expressions of appreciation, not only for the gift, but for the contribution that the Harris Matrix and its associated concepts have made to the practice and science of archaeology, apparently around the world.

Via the young technology of the internet, it is hoped that the principles of archaeological stratigraphy, as elucidated over a generation ago, can now reach every corner of the globe. The older technologies of excavation and recording can be replaced by methods through which the past can be better served and preserved, as we destroy its physical remains in our scientific work.

The author's Principles of Archaeological Stratigraphy (Academic Press, 2nd ed 1989) can be downloaded for free at www.harrismatrix.com. EC Harris is executive director of the Bermuda Maritime Museum. In 2000 he received the MBE for services to architectural, maritime and cultural history in Bermuda.

CBA web:

British Archaeology

February 2001
April 2001
June 2001
August 2001
October 2001
December 2001
February 2002
April 2002
June 2002
August 2002
October 2002
December 2002
March 2003
May 2003
July 2003
September 2003
November 2003
January 2004
March 2004
May 2004
July 2004
September 2004
November 2004
Jan/Feb 2005
Mar/Apr 2005
May/Jun 2005
Jul/Aug 2005
Sep/Oct 2005
Nov/Dec 2005
Jan/Feb 2006
Mar/Apr 2006
May/Jun 2006
Jul/Aug 2006
Sep/Oct 2006
Nov/Dec 2006
Jan/Feb 2007
Mar/Apr 2007
May/Jun 2007
Jul/Aug 2007
Sep/Oct 2007
Nov/Dec 2007
Jan/Feb 2008
Mar/Apr 2008
May/Jun 2008
Jul/Aug 2008
Sep/Oct 2008
Nov/Dec 2008
Jan/Feb 2009
Mar/Apr 2009
May/Jun 2009
Jul/Aug 2009
Sep/Oct 2009
Nov/Dec 2009
Jan/Feb 2010
Mar/Apr 2010
May/Jun 2010
Jul/Aug 2010
Sep/Oct 2010

CBA Briefing

Fieldwork
CBA Network
Conferences
Courses & lectures
Grants & awards
Noticeboard

British

Archaeology

The voice of archaeology in Britain and beyond

Issue 103

Nov / Dec 2008

Contents

news

features

on the web

letters

CBA correspondent

features

35 years of the Matrix

CBA web:

British Archaeology

CBA Briefing

CBA homepage