Bryan Sykes and Catherine Irven, Surnames and the Y Chromosome, 2000

This conclusion by Bryan Sykes and his colleague achieved widespread publicity in the media, arousing interest generally and having an enormous impact on genealogists. His conclusion, based on a 1999 DNA experiment on sixty-one volunteer males named Sykes, inferred that men who bear this surname are likely to share a common progenitor. This revelation suddenly brought attention to the theory that many surnames have single-family origins. The earliest history of the surname suggests that in the late 1200s the progenitor lived in an area to the west of Huddersfield identified later in this article as Flockton/Saddleworth. The new scientific evidence meant that families in several different parts of this region could now be linked, and their migration routes traced over a difficult period of more than 200 years.

The idea of single-origin surnames has been advanced often enough in the past but it had seemed incapable of proof by traditional genealogical methods, so this scientific contribution to the argument excited both genealogists and surname specialists. There would have been less surprise if similar research had suggested that all males named Thistlethwaite share a common paternal ancestor, but Sykes has been thought of as a much less distinctive surname (derived from a word meaning “stream” or “ditch”), and likely therefore to have had several origins.

For almost forty years I have argued that many surnames are likely to have had single origins, even some of those commonly thought to have multiple origins. I expressed that view as long ago as 1973 in Yorkshire: West Riding (Phillimore, 1973), the first volume in the English Surnames Series edited by Dr. David Postles. I have to say, though, that I shared the more general view that Sykes probably had a number of distinctive origins, due to its frequent appearance in early records throughout widely scattered parts of west Yorkshire and east Lancashire. I was, of course, fully aware of the spectacular expansion of Sykes in and around the village of Slaithwaite in the Colne valley after 1480, and I certainly considered these families to share a common origin, but the earlier history of the name still seemed to point to more than one source.

Bryan Sykes’s brief account of the experiment brought far more publicity than academic articles usually enjoy and, not surprisingly, there were some skeptics, although these tended to be family and local historians rather than geneticists. One of the main reservations had to do with the relatively small size of the sample, for there are approximately 10,000 registered voters named Sykes, and DNA was successfully extracted from only 48 individuals. Moreover, the volunteer families had been selected “at random” from Yorkshire, Lancashire, and Cheshire, the counties with “the highest residential concentrations” of the surname, and it was felt by many that the test should have included at least some individuals from outlying areas.

Neither of these objections appears to have troubled geneticists unduly. Dr. Thomas H. Roderick, a staff member of the Center for Human Genetics, thought that individuals who were adopted or who suspected “a non-paternity event in their Y line” would have been unlikely to respond to the request for a DNA sample. Even if that were true, the results would only have understated the number of men called Sykes who owed their surname to adoption or illegitimacy. They would not have pointed to an alternative origin for the surname.

Mark Jobling, a Senior Research Fellow in the Department of Genetics in the University of Leicester, commented on Bryan Sykes’s use of only four micro-satellites in his experiment and was surprised that these could be so “Sykes-specific.” However, the haplotype (a set of genetic determinants located on a single chromosome) in this case was  “common in the Baltic States but rare or absent elsewhere,” --making it so distinctive that Jobling conceded that it was probably unusual enough in Britain “to be resolved by only a few markers.” His other cause for surprise had to do with the surname’s “topographical derivation” and this was really much more on the lines of the doubts expressed by surname specialists.

The results of the experiment carried out by Bryan Sykes are not in question but his conclusions have opened up so many possibilities for genealogists that it seems worth examining the early history of the Sykes surname, if only to address the objections about the “sample.” It is at least worth looking more closely at the name’s origin and meaning, and at its distribution in the fourteenth century, when surnames were stabilizing and before the dramatic expansion in
Flockton/Saddleworth would cloud the issues.

The first point to take into account is the limited and regional use of the word “syke” in place-names. The evidence reveals that its distribution is much tighter than surname dictionaries have suggested, for it does not occur outside the northern counties. Even there its range is limited and elsewhere it is found as “sitch,” even in west Lancashire, drawing our attention to the similarities with “dyke” and “ditch.” Etymologists concede that the variation between “ditch” and “dyke” has not been fully explained but note that the frequency of “dyke” and “syke” in areas that were formerly part of the Danelaw may point to Scandinavian influence. Whatever the explanation the fact is that Sykes derived from a place-name that had an extremely limited distribution, so we would expect the surname also to have a regional distribution. Indeed it was identified by H.B. Guppy, in The Homes of English Family Names (1890), as a name largely confined to Lincolnshire and the West Riding of Yorkshire.

In the West Riding poll tax of 1379, for example, there were men called “de Syke” or “de Sykes” in no fewer than ten different villages. Richard de Sykes lived in the mountainous northwest of the county, in Newton in Bolland, whilst sixty-six miles to the southeast, in a landscape of fens, was William de Syke of Fishlake. Between these two extremes were families in Bradford and parts of Airedale (Baildon, Horsforth, and Bingley) and also in Emley and Saddleworth, the area where the surname later expanded so dramatically. A Sykes family taxed in Rochdale, Lancashire, in 1379 could be included in the same area. The first recorded examples of Sykes are in Flockton, near Emley, in the late 1200s, and the name was probably hereditary from that date. For those familiar with surname distribution in this period it is difficult to accept the idea that all the families taxed in 1379 would have been related.

However we are not looking at ten or eleven possible origins. The families taxed in Yorkshire and Lancashire in 1379 may have been scattered over a wide area but within that general distribution some were living close enough together to be related. This would still leave us with at least four significant concentrations: Ribblesdale (two families), Flockton/Saddleworth (two families), Bradford/Airedale (four families), and Fishlake (two families). Of course it is conceivable that some of these names were non-hereditary by-names and that others may have become extinct in the male line soon afterwards. A search of fifteenth- and sixteenth-century records throws some light on these possibilities. For example, the printed subsidy rolls of 1545 show concentrations of Sykes in Flockton/Saddleworth and in Bradford/Airedale, but none in Ribblesdale, whilst parish registers and will indexes suggest that Sykes could still be found in good numbers in and around Fishlake, including neighbouring parts of Nottinghamshire. The overall inference is that the name had survived in three of the localities where it had been recorded in 1379 and this distribution is remarkably similar to that commented on by Guppy in 1890.

If the volunteer families tested by Bryan Sykes were drawn more or less equally from these three areas, even the most skeptical genealogist would have to accept that those who bear the surname share a common ancestor, however unlikely that might at first seem. It would also mean that Sykes has a single place-name origin, in which case Sykehouse in Fishlake would be the obvious source, a derivation with interesting implications for the surname’s precise meaning.