Have you encountered a variation in your surname and dismissed the possibility of an ancestral connection? Many of us have made that mistake. And it seems to happen even more frequently with Irish surnames, in part because of the prefixes O and Mac. I'd previously understood that Mac meant 'son' (or 'son of') and O' meant 'grandson of'. I'd also read that 'Mac' indicates a Scottish name while 'Mc' is Irish.

But are these true statements? I just read an article about surname variations, written by Sean J. Murphy and published in a 2013 issue of Irish Roots magazine, which answers those questions and others. The information for this blog posting comes from Murphy's article. Surnames in Europe evolved during the 10th and 11th centuries.

They were derived from four main categories: 1) an ancestor's name, 2) a placename, 3) an occupational name, or 4) a descriptive name. Most Irish/Gaelic surnames had ancestral origins, exemplified by the prevalence of O and Mac prefixes. Further, some surnames are polygenetic, i.e., arising independently in different places. Other names are monogenetic, i.e., coming from a single point of origin. Based on what he calls the 'mono-poly' distinction, Murphy dismisses the 'romantic pitfall' that the Irish were divided into clans. Instead, like Edward MacLysaght and other scholars, Murphy favors use of the word 'sept'. He puts forward his own surname, Murphy (Gaelic Ó Murchadha), which is one of the most common in Ireland. He says Murphy refers to unrelated septs which originated in Cork, Roscommon and Wexford. Other Gaelic surnames are similarly polygenetic. For example, the name O'Kelly/Ó Ceallaigh was historically located in Galway, Derry, Laois, Meath and even Wicklow; O'Connor/Ó Conchobhair was the name of unrelated septs in Connacht, Kerry and Clare.

An exception: The Munster sept of O'Sullivan/Ó Suileabhain appears to be monogenetic. DNA analyses will continue to help define surnames. So what about O and Mac? The Gaelic prefix 'O' originally meant grandson of the person whose name followed while 'Mac' meant son. As time passed both prefixes came to have a more general meaning, i.e., descended from named individuals. The surname O'Brien was first used by grandsons of Brian Boru. However people frequently adopted the name of overlords so not all O'Brien's are Brian Boru descendants. And some O'Byrnes and Bryans may be truly related to O'Briens. Again, further DNA analysis will shed light on genetic relationships. Murphy says use of the apostrophe O' resulted from an English error confusing the abbreviation of 'of', e.g. Glen O' the Downs. And the prefix M' in older records may be a similar abbreviation for Mac.

The Scots used the prefix Mac in forming surnames and rarely used O; however, that doesn't mean all or most Mac surnames in Ireland originated in Scotland. Murphy says this is particularly true in Ulster and cites examples of the names MacAuley and MacKinney. Murphy dismissed the view that O and Mac prefixes were discouraged/prevented under English rule in Ireland. He found no English laws enacted to prohibit use of the prefixes. Nonetheless, Irish surnames were anglicized, and later generations have subsequently restored disused prefixes. He completed a survey of telephone directory entries in the 1990s and compared it to Robert E. Matheson's study of 1890 birth registrations. A Survey of Irish Surnames 1992-1997 (by Sean J. Murphy)

Surname	% using prefix 1990s	% using prefix 1890
(O) Kelly	3	1
(O) Connor	89	32
(O) Sullivan	89	14
(O) Byrne	5	1
(O) Brien	98	67
(O) Neill	92	63
(Mc) Carthy	97	85
(Mc) Laughlin	92	87
(Mc) Mahon	71	74
Maguire/McGuire	100	100

Note many O'Connors and O'Sullivans have restored use of the prefix O while Kellys and Byrnes have not. Interestingly the Mac prefix was dropped less frequently than O - so the rate of re-adoption is small. Murphy concludes his article with this recommendation:

...the genealogist should search records and databases firstly under the received form of a surname, with or without an O or Mac prefix or spelled in a certain way. If results are negative, the search should be broadened to include variants of the surname, which will sometimes be thrown up by better designed online databases or else can be identified by reference to works such as MacLysaght's The Surnames of Ireland or Mathesons' Surnames in Ireland. Which leads us back to the proposition that we genealogists should not have an inflexible approach to 'right' or 'wrong' spellings of our surnames.