The following is from the book “Surnames, DNA, & Family History”; by George Redmonds, Turi King and David Hey.

Son of Robert

 Another difficult task for non-specialists is to discover what type of origin a name has and how it relates to similar names with much the same meaning.  Some of the issues emerge in a study of three surnames that all mean ‘son of Robert’.

Robinson, with a total of 95,495, is an excellent example of a multi-origin name, but it is characteristic only of the northern counties of England and is not generally popular.   In 1881, even after centuries of internal migration, Robinson was still very uncommon in all the counties in south-west England and was rare in both Wales and Scotland, whereas in Yorkshire over 20,000 people bore this name.  Of course, the totals for the biggest or most populous counties can be misleading, and this is where the relative concentration of the name comes into question.  These statistics, which are also provided on the Archer CD, show how significant a name Robinson was in the smaller or largely rural counties, notably in Westmorland, Cumberland, and the North Riding of Yorkshire, and in Lincolnshire, which was virtually the southern limit of the surname.

Robertson has much the same meaning as Robinson but its history is totally different.  In 1881 it was the 56th most frequent surname in Britain with a total of 50,256.  Most Robertsons had their home in Scotland, where the name owes much of its importance to the Clan Donnachie, having been assumed as a family name by William, who became chief of Struan in 1509.  Subsequently, it passed into general use among the members of the clan.  It was still very prolific in Perthshire in 1881 with a total of 4,179, smaller than the total of7,644 in Lanarkshire, but a much higher percentage of the population.  Another high concentration was found in the Shetland Isles, where Robertson is likely to have had a quite different history.  These figures can be contrasted with those for England, where Robertson was generally very uncommon.  Understandably, it had a significant presence in counties such as Northumberland and Durham, and in one or two areas of high population, but in much of the southern half of England it was virtually unknown.  In Huntingdonshire, for instance, just two people were called Robertson.

A third name to be considered alongside Robinson and Robertson is Robson, again meaning ‘son of Robert’.  In 1881 the Robsons numbered 19,730 and the name was 190th in order of popularity.  In the sixteenth-century history of the Borders, the ‘Debatable Lands’ between England and Scotland, the Robsons were one of the four great clans or ‘graynes’ who dominated the North Tyne.  The name is still markedly more popular there than anywhere else and in 1881 more than half of all the Robsons were still living in Northumberland and County Durham.  Elsewhere, the name was not at all common and that is again particularly true of the South-west, for Cornwall had just two Robsons and Dorset only six.

Clearly, these three surnames have to be treated as distinctive and, because two of them are so closely linked with the clans and extended families of the North, questions remain unanswered about the nature of their origins.  There is uncertainty too about the names Robison and Robeson, which were identified by Black as Scottish variants of Robertson.  Theoretically, they might also be variants of Robinson and Robson but the numbers are in any case relatively small.  The three surnames have multiple origins and they use -son as a suffix.  This type of formation has been shown to be characteristic of the northern half of Britain and all three have their equivalents in other regions: Robins or Robbins, Robert and Roberts, Rob and Robb.  This is not an exhaustive list but it illustrates the diverse ways in which ‘son of Robert’ could develop in different parts of the country.  Such matters should be borne in mind when the popularity of a name is discussed, but they should not prevent those discussions taking place, for the statistical evidence can be used to great advantage by demographers, geneticists, and researchers in other fields.  The benefits for local and family historians lie mostly in the information available about particular names, but the relative popularity of British surnames is of more general interest.