Bringing it all together

At present I am commencing a study of the surnames listed in the Portsmouth Burgess Rolls of 1900/01. Will Portsmouth with its strong naval tradition be a melting pot of UK names? The Burgess Rolls will reveal only a section of the population, but that section will be one that has put down roots and consequently the surnames are less likely to be held by transients. Ideally I should compare the results with a sample take from the 1881 Portsmouth census.

This will be a testbed of the techniques already discussed to reveal:-

  • S/N - the degree of surname diversity in Portsmouth
  • The percentage of singletons
  • A cumulative frequency graph
  • How the incidence of leading names compare with the national average
  • The incidence of local surnames
  • The incidence of surnames that have subsequently become extinct in the UK
  • Also, I am interested in how the initial results of the start ward are affected as the area grows larger by feeding in the results from more wards. Will more previous singletons merge, than are bought in by the new ward? How will S/N change? If a name predominates in one or two wards, does that suggest a localised kinship?

    Initial results

    As these are the results for the first two wards, they are meant just to illustrate the techniques, rather than being definitive. They will change as I include neighbouring wards.

    There is a remarkable consistency between the figures for the two wards. The number of surnames is perhaps 10% larger due to spelling inconsistencies, e.g. Cleal and Cleall are treated as individual names.

    Mile End St Mary's
    S 1329 1309
    N 2291 2237
    S/N 0.58 0.57
    names householders names% names householders names%
    Ones 917 917 69% 926 926 70%
    Twos 222 444 17% 208 416 16%
    Threes 82 246 6% 77 231 6%
    Fours 34 136 2.6% 27 108 2.1%
    Fives 25 125 1.9% 17 85 1.3%
    Sixes 14 84 1.0% 12 72 0.9%

    480 surnames are common to both Mile End and St Mary's Ward - of which 220 are singletons. In other words, over 60 % of surnames are different in the contiguous wards (whose centres are only a couple of miles apart). The combining of the 2 wards results in a larger influx of singletons, than the number deleted by merger. Deleted: 220 : Influx 706. Some of these 706 unite with multiples in the core ward. Result: the % of singletons in the combined ward declines to 64%

    Ones 1385 64%
    Twos 262 12%
    Threes 153 7%
    Fours 91 4%
    Fives 46 2%
    Sixes 23 1%

    Core Ward: Mile End

    Graph 1
    Cumulative Frequency graph of the 1329 names.
    Note the straight line where the singletons commence.

    Graph 2
    The top names in Mile End compared to 1853 England and Wales equivalents.
    Welsh names seem to be under-represented in this Ward. This may be compensated in other wards. Portsmouth has had an association with South Wales, e.g. migration from Pembroke Docks.

    Graph 3
    The baseline is the leading names of England and Wales in 1853, against which the ward is being compared (assuming a stability in surname percentages in the intervening 50 years).
    Columns = Mile End 1900.
    Again, Welsh names are under-represented whilst the name White is present in almost double its national percentage.

    Possible Extinctions:

    Local Names (25 mile radius):
    Cawte, Edney, Oakshott, Privett, Stares, Brading, Burt, Damerum, Mengham,

    Sheffield 1841 : A Comparison

    David Hey in A History of Sheffield devotes a few pages to the surnames of the city. Incidentally, this is the only instance that I know of the study of the surnames of an urban area, rather than a county or hundred. More please.

    In the 1841 Sheffield census, he finds that the following are the most common surnames:

    1841 Census Sheffield : Rank: Name: Occurrences
    1 Smith 1624 9 Johnson 580 17 Green 521
    2 Taylor 1022 10 Ward 573 18 Marshall 503
    3 Wilson 925 11 Thompson 572 19 White 486
    4 Walker 678 12 Rodgers 571 20 Lee 485
    5 Hall 660 13 Brown 567 21 Barker 466
    6 Turner 628 14 Haigh 563 22 Clark 452
    7 Shaw 605 15 Jackson 541 23 Roberts 450
    8 Wright 586 16 Parkin 531 24 Robinson 443

    An unrealistic comparison, I know, not comparing like-with-like timewise, but indulge me

    • The surname profile is markedly different to that of Portsmouth (Mile End) 1900.
    • Only 3 names are common to both - Smith, Taylor and Brown.
    • Note the complete absence of Jones (does not even reach the top 50 names). Welsh migration to Sheffield came later.
    • Wilson and Walker are northern names, and would be expected to have a high occurrence, as is Parkin.
    • Johnson and Haigh suggest migration from Scotland.

    (Source: David Hey A History of Sheffield Carnegie Publishing Ltd, 2nd Revised edition edition, 2005.)

    R J Johnston

    A more contemporary post-war analysis has been developed by R J Johnston, who examised the kinship factor as a deterrent to migration through a study of the common surname groups of the Yorkshire Nidderdale in the period 1951-1961 through the use of the electoral rolls. As a check on the localisation of the surnames under consideration, as a comparison, he counted the number of entries in the following telephone directories, and adjusted the resulting count with a weighting factor.

    The following is a much pared version of his results, to concentrate on urban areas in a general north to south swathe, and the less localised of his surnames.

    Name Middles-
    Leeds Leicester London Ports-
    Exeter Mean
    Ashby 26 38 68 54 32 46 44
    Barker 722 571 439 200 164 168 377
    Clough 64 143 17 9 7 14 42
    Fawcett 268 202 32 17 28 20 95
    Harris 232 357 756 704 857 114 503
    Hudson 352 619 207 130 15 114 240
    Pratt 225 155 293 84 207 177 190
    Simpson 768 524 463 235 285 257 422
    Smith 2857 3024 4219 2000 2857 1857 2802
    Thompson 1857 1083 732 416 500 457 841

    The weighted counts that are above the median are highlighted. Nearly all the names seem to exhibit a north or south bias. The surnames Ashby and Smith appear to be names predominantly found in the Midlands. (The repeat of 2857 in the Smith row is not a typo). It will be interesting to see if the high occurrence of Harris is repeated in my Portsmouth study.

    (Source: Johnston, R. J. Resistance to Migration and the Mover/Stayer Dichotomy: Aspects of Kinship and Population Stability in an English Rural Area. in Geografiska Annaler, (1971) 53B:16-27.)

    Further Work

    Possible group projects might be:

    • A examination of surname profiles for differing geographic communities in the 1881 census
    • The investigation of name loss or creation in the twentieth-century through a comparative study of a ward electoral register for selected periods 1930-1999
    • A surname profile derived from Hearth Tax lists, or a local census.

    Even more intricate work could be done on the numbers of type of surname (if identifiable as locative, personal etc), and percentage influx of new names.