Journal of Historical Geography, 10, 1 (1984) 1-14


Migration fields in a traditional Highland community, 1350-1850

Alan G. Macpherson

Using information on marriages from a clan genealogy composed in 1705 and from parish registers after 1775, a network of social relations within the Clann Mhuirich has been constructed over a three-and-a-half century period. An examination of marriage records and records of' the residences of' baptizing families throws light on the means whereby social relationships are formed and maintained over space through time in a Scottish highland community before the mid-nineteenth century. A basic characteristic of a Highland clan was its agnatic structure, binding men and women related to each other by patrilineal descent, often bearing the same surname. The territory occupied by a clan became segmented by a process of agnatic migration as clansmen acquired land through marriage. Another characteristic of clan organization was a tendency towards high levels of clan endogamy. Endogamous marriages reinforced the agnatic structure and maintained continuity of possession of a clan's territory.

Those ot the same name and clan look on themselves all as gentlemen and
bretheren, and the chief as the comon father or parent from whom they all
come and count their liniall descent so that they fight . . . as children of the
same family joined in regiaments togither ...

John Erskine, Earl of Mar, 1726 [1]

Inquiry into the geography of past social contacts must, inevitably, begin with a search for adequate indicators and if the kind of contact sought is to be persistent and pervasive, the indicators available are likely to be few in number. One of the most readily available -- for recent centuries at least -- is marriage migration.

      The traditional society of the Scottish Highlands with which this paper is concerned differed considerably in its basic structure, and therefore in much of its spatial behaviour, from what has come to be regarded as the mainstream of British society. If one is tempted to regard traditional Gaelic society as aberrant one should remember that its peculiar social structure and cultural norms were dominant over large areas of the British Isles prior to the Tudor reconquest of Ireland and continued important in the Highlands and Islands until after the Napoleonic Wars.

      Earlier work, using the same documentary sources as in this paper, has established that the social structure of the Highlands was one of' cantonal communities composed of a number of distinct clans or extended kin-groups, the dominant clan of one district usually extending in particular lineages (sliochdan) into adjacent districts where these then assumed the position of a dependent or minor clan (clann) in that community. It was this interlocking cantonal and community structure that was referred to in the disturbed years of the early

0305 7488/84/010001 + 14 $03.00/0

© 1984 Academic Press Inc. (London) Ltd.


eighteenth century as "the clan system"; Its fundamental units were the local lineages associated with particular farms or groups of farms in which rights of ancient possession were acquired and asserted through conjoint tenure and the practice of runrig.

      Historians have usually been attracted to the para-military function of the Highland clan, but this -- in an anthropological sense -- was merely a secondary characteristic. The primary characteristic of the clan was its agnatic structure, that is, it was composed of men and women related to each other by patrilineal descent from a common (and often eponymous) ancestor; with the passage of time, through a number of generations the original kin-group was subjected to a process of segmentation as male members succeeded in establishing lineages in farms in which new rights of possession were acquired for their clan. The territoriality of each clan was, therefore, an idiosyncratic expression of its segmented structure resulting from a spatial process which may be referred to as agnatic migration.

      Figure 1 shows the pattern of agnatic migration and the resulting distribution of lineages of a clan which, to avoid confusion, shall be referred to by its original name in Gaelic, the Clann Mhuirich. [2] It represents out-migration through some two-and-a-half centuries (1450-1700), initiated from three farm towns in the parish of Laggan (including the headwaters of the Spey in the old Lordship of


Badenoch, south eastern Inverness-shire) and continued by secondary migration from farms inside and outside the parish in which rights were acquired in earlier phases of the process. In some instances secondary migration involved return to primary centres, particularly when rights were lost or abandoned at secondary centres. The source of the information is a manuscript genealogy of this one, albeit dominant, kin-group, compiled in 1705 and embracing some 957 members over twelve generations; it also includes references to over 200 exogamous marriages in which fathers-in-law of other kin-groups are mentioned by name and territorial designation. The document, corroborated by reference to contemporary records and independent traditional statements in which individuals are often identified by a combination of patronymic and territorial[3] association, has been thoroughly authenticated.

      In a positivist sense, Figure 1 represents not only the spatial structure of that part of the Clann Mhuirich originating in Laggan as it existed in 1705, but also the process which gave rise to the structure. In an idealist sense, the figure represents the perception of his clan entertained by the compiler of the manuscript genealogy, and his participation in the social interaction sustained among scattered lineages of the clan. The compiler was a political activist of his time, concerned with the para-military function of the clan system as a whole, and with the solidarity and cohesiveness of his own clan in particular; it can be assumed, therefore, that the social contact sustained within the clan was of a high order, capable of conversion into political and para-military action. The figure is therefore one mode of the social interaction field for one particular kin-group. If the genealogy betrays lacunae in the information, it simply means that, in 1705, the missing lineages were no longer part of the interaction field centered in Laggan.

      In the case of the Clann Mhuirich kin-group at the beginning of the eighteenth century, social contact among its member lineages can be summarized as, first, an intense concentration within the parish of origin and in the two downstream parishes of Kingussie and Alvie which comprised the lower parts of Badenoch; secondly, a more open pattern in the valleys of the Findhorn and Nairn and in the vicinity of Inverness, produced by early moves from Laggan (1450-1550) from which return to Badenoch occurred in one or two instances in the early seventeenth century, indicating sustained contact and an acceptance of responsibility by senior lineages for junior lineages in process of losing possession of their land-rights; and thirdly, three more detached areas of contact following acquisition of lands in the Braes of Lochaber, in Banffshire, and in Lowland Angus between Dundee and Montrose.

      Earlier work has shown that agnatic social structure in the Highland clan was accompanied by another primary characteristic, a socially-directed tendency towards high levels of clan endogamy. The actual percentages of men and women at risk conforming to this direction varied from clan to clan, but once the practice had become established in a newly emergent clan, say by the fifth generation, the percentage of endogamous marriages tended to climb to a particular level and remain there. In the case of the Clann Mhuirich this level was between 60 and 65 pct irrespective of the proportion of the community provided by the clan in any particular generation. A tendency towards endogamy was probably, in essence, an expression of a need to retain and circulate wealth within a clan, and perhaps to preserve rights of possession. Where sisters and daughters of leading men married into the clan the political cohesion of the whole kin-group was undoubtedly strengthened. Marriage migration, therefore, included an


endogamous component consisting of individuals -- mainly, but not exclusively, women -- moving from one lineage location within the agnatic structure to another and, with respect to lineage locations in Laggan, moving across the agnatic structure in and out of the parish. The endogamous component in marriage migration, therefore, represented a reinforcement of the basic agnatic structure and, in terms of process, grew along with it. More than any other factor, it probably accounted for the, strength and intensity of the interaction within the clan. Its spatial aspect can be seen as a second mode of the social interaction field centred in the parish of Laggan and specific to the Clann Mhuirich kin-group living there.

      Figure 2 shows Clann Mhuirich spouses moving within and out of Laggan at marriage; Figure 3 shows the reciprocal counter-movement into Laggan. Together they indicate that the social interaction field was intensified along the upper Spey in Badenoch, while contact was maintained with geographically detached lineages in Lochaber, Strathnairn and Lowland Angus. Emphasis on patrilineal descent, however, has introduced bias into the source document, and some lineages are recorded without wives and, more damagingly, without sisters. Figures 2 and 3 incorporate this bias and therefore represent minimum statements.


      If marriage within the clan reflected and helped to sustain intense social interaction within the agnatic kin-group, marriage outside the clan undoubtedly represented broader economic and political interests. A third characteristic of the primary behaviour of the Highland clan was the tendency for leading men of important and influential lineages to marry exogamously while most of their sisters and younger brothers married endogamously. Indeed, of 15 marriages involving 11 successive chiefs in ten generations of the Clann Mhuirich only one -- the fifth generation --was endogamous; whereas of 11 marriages recorded for brothers and sisters of 4 chiefs in generations VIII to X (roughly the seventeenth century) 7 (64 pct) were endogamous. Figure 4 is a fairly full statement of the inmigration at marriage of women belonging to clans and families with whom men of Clann Mhuirich in Laggan formed alliances of various kinds from time to time. Figure 5 represents reciprocal out-migration of Clann Mhuirich women from Laggan to form exogamous unions elsewhere.

      Examination of 38 exogamous marriages in generations VII to XI indicates that:
(a) only 8 involved alliances within the parish;
(b) these 8 were exclusively with three local lineages of two other clans prominent in the community: the Sliochd Iain Leith and Sliochd Dhomhnaill


Ghlais of the Maclntoshes and the Sliochd Iain Duibh of the MacDonalds of Brae-Lochaber, both dominant clans in adjacent districts:
(c) marriages with Maclntoshes (12) and with MacDonalds of Brae-Lochaber, Glengarry and Glencoe (6 ) -- including 8 already mentioned -- constituted almost 50% of the contact;
(d) 3 were with members of the immediate family of chiefs of these clans;
(e) 23 were with families of Highland tacksmen belonging to a dozen different clans;
(f) 7 were with families of lairds or feuars of small estates;
(g) one was with the family of a parish minister;
(h) two were with merchant provosts of Inverness;
(i) none were with members of the Scottish aristocracy; only two were with artisans: a miller and a smith.

We are clearly dealing with a rural elite interacting over a fairly extensive, but still well defined area, within which the agnatic structure of the Clann Mhuirich was confined. The spatial limitations of the social arena depicted in Figures 2 to 5, taken together are quite impressive: very few contacts southward into Highland Perthshire, and none into the Lowlands, except for one geographically anomalous endogamous union; no contacts with the Southwest Highlands of


Argyllshire or with the Hebrides, and virtually none beyond the Great Glen; and -- surprisingly -- none whatever across the dissected plateaux to the east. The agnatic migration process of Figure 1 can be seen as an extension within this arena, in which the kin-group from Laggan was seizing opportunities and overcoming resistance from rival kin-groups more easily than it could outside the arena. It is reasonable to suppose that its success was related to earlier states of the interaction field reflected by the heiratskreis for exogamy; Figures 4 and 5 therefore represent a third mode of the interaction field, probably related to political and economic interests in the first instance.

      At this point it is worth pausing to consider the full implications of migration at marriage as an indicator of persistent and pervasive social contact or interaction. The movement of a woman from parental to husband's hearth at marriage was preceded in a traditional society of the kind under discussion by courting expeditions on the part of the prospective husband, often accompanied by one or two male companions of his peer group if the distance involved was relatively great. Courting visits were preceded by social visits of a more general nature, often involving parties of young men, the excuse for such visits lying in still earlier social, political, or economic contacts, including such customs as fosterage. Moreover, for a woman moving more than a few miles at marriage there was generally no return, even for purposes of social visiting; marriage migration


carried with it a sense of exile which became very real as distance increased. Evidence for this is slight and must be interpreted cautiously; but for what it is worth two cases involving daughters of clan chiefs can be given: about 1685 Sileas MacDonald (1660--1729), a daughter of Archibald MacDonald of Keppoch, moved at marriage from the foot of Glen Roy in Brae-Lochaber to upper Banffshire, a distance of some 120 kilometres, and seems to have spent the rest of her life there. She expressed her sense of isolation in poems in which she describes herself plying visiting Lochaber-men with questions concerning the affairs of families belonging to her clan; her Banffshire husband, on the other hand, visited Lochaber before and after their marriage, basically on political business.[4] The second case concerns the well known song of exile "Lochaber No More", which has acquired the connotation of the trans-atlantic migration, but which was written by Jean Cameron of Lochiel on the occasion of her migration at marriage in 1704 from Achnacarry on Loch Lochy-side to Nuide in Badenoch, a distance of some 80 kilometres; later she moved with her husband and young family into Laggan, and there is a wealth of documentation on visits by male kinsmen on both sides of the marriage, travelling between Laggan and Lochaber into the 1740s, long after her death. In migrating to her husband's home, the Highland woman was initiating a counter-movement of male kinsmen for whom her destination was often a new locus for social interaction. Her points of origin and destination, in fact, defined a specific channel for such contact, activated or reactivated prior to her marriage, sustained long after it, and waxing and waning in the direction of maximum movement perhaps with the seasons or the passage of years. The social interaction field associated with a traditional society can be thought of as consisting of a multiplicity of such channels.

      The material presented so far has depended upon three primary characteristics of a clan-based order of society, and the order of presentation might suggest that agnatic migration was a pre-requisite, absolutely necessary for any endogamous marriage migration, and probably necessary for any significant level of exogamous marriage migration. It raises the question of antecedents in time for interaction fields developed between 1450 and 1700. There is evidence, however, that traditional society in the Highlands was also characterized by occasional assumption of matrilocal residence at marriage: that is, that in perhaps one marriage in six or seven it was the man who migrated, to be admitted to residence among his wife's and father-in-law's kin-group or clan. Where dominant clans were concerned, Highland genealogies often indicate that the lineage descended from such a marriage "counted its genealogy" among lineages of the clan among whom it resided, rather than among those of its own clan; this undoubtedly accounts for the incomplete nature of most Highland genealogies. In the case of the Clann Mhuirich and the 1705 manuscript, a number of gaps can be identified, and invariably refer to lineages living among the Clann Raghnaill MacDonalds of Brae-Lochaber, the Maclntoshes of Strathdearn and Strathnairn, and minor baronial families in the lowland between Nairn and Inverness. In some of these cases matrilocal residence is indicated, although there is uncertainty about the actual order of events: whether the man had moved before or after marriage.

      Matrilocal residence, as one would expect, was also a consequence of endogamous marriage. Its inclusion as an explanatory element in the mechanisms of social contact, therefore prompts examination of the apparently exogamous unions of the newly emergent Clann Mhuirich, that is, marriages in generations I to IV in the fourteenth and early fifteenth centuries. Of 17 marriages recorded for


11 men and 4 women of the incipient kin-group, no fewer than 11 (65 pct) were with clans for which there exist early traditions of agnatic kinship with the Clann Mhuirich. It could be argued tentatively that these marriages, while exogamous to the Clann Mhuirich, were endogamous within an ancestral kin-group out of which the Clann Mhuirich, the MacLeans, MacIntoshes, Camerons, MacBeans, MacNivens and Davidsons all emerged in the thirteenth and fourteenth centuries. The argument, tentative as its basis must remain, raises matrilocal residence after marriage to the position of a fourth primary characteristic of traditional Highland society; as it made a significant contribution to agnatic migration in both the endogamous and exogamous fields of interaction it effectively removes agnatic migration from the role of prior requisite and reduces it to just one of the components in the field dynamics. The theoretical relationships can be shown as follows:

      The contention that agnatic and marriage migration reflect a strong and persistent field of social interaction can best be supported by content analysis of relevant muniment collections. The index to the Cluny Papers[5] lists some 450 documents focused upon Clann Mhuirich affairs between 1571 and 1705; the index to the MacKintosh Muniments[6] lists some 21 documents between 1564 and 1705 in which men of the Clann Mhuirich from Laggan are associated with the affairs of the Laird of MacIntosh. The affairs in each case consist largely of transactions involving money, land, and women -- all three in the case of marriage contracts. Whenever Clann Mhuirich men from Laggan are documented as acting as principals or witnesses in these transactions all other individuals involved were extracted by territorial designation or residence: some 340 items referring to some ninety different places shown in Figure 6. The agnatic structure is prominent in the pattern, and the rest of the field generally corroborates the orientations, intensities and limits depicted in Figures 4 and 5.

      So far, attention in this paper has been restricted to one kin-group -- albeit a dominant kin-group in its parish of origin -- and to the period ending in 1705. In order to test the validity of the model for the whole community of Laggan and for a later period, historical demographic methods were resorted to. Family reconstitution of the baptismal and marriage register of the parish of Laggan for the period 1775-1854 has demonstrated that the whole community was, and remained, strongly agnatic in its internal organization and territoriality; it continued to favour clan endogamy and to observe a discriminatory or blocked practice of exogamy, and to accept a considerable amount of matrilocal residence. Only after the Napoleonic Wars did the high level of endogamy in all

------------------------------------------------------------------ 9--------------------------------------------------------------

kin-groups decline towards a position of random choice, approaching the null hypothesis for the whole community. At the same time, and probably related to this change, selective migration of families belonging to the dominant clans in the community -- including the Clann Mhuirich -- led to an increase in matrilocal residence: men of minor kin-groups within the parish and men from more distant points of origin began to replace brothers-in-law, initiating a more general process of change which terminated in the mid-twentieth century with the virtual disappearance of the traditional kin-groups.

      The schema below summarizes the data-set involved in the analysis. Baptism of children is assumed to be an indication of residence in the community (70 pct of the reconstitutions). Baptizing families with a record of marriage are assumed to be reasonably representative of the community as a whole, largely on the basis of identical surnames, combinations of surnames, and territorial associations of surnames. Of the latter group, 28 pct included a spouse from outside the parish, and it is this sub-set that can be used to define the in-migration heiratskreis for Laggan for the last quarter of the eighteenth century and first half of the nineteenth century, as shown in Figure 7. Record of marriage without consequent baptisms is assumed to be largely indicative of out-migration, but only the sub-set involving a spouse from outside the parish (usually the man) can be used to define the corresponding heiratskreis. (Figure 8) The two sub-sets together represent the social interaction field of the community of Laggan between 1775 and 1854, insofar as it can be delineated by this method.



      Correspondence to the field for the Clann Mhuirich prior to 1705 is strong: the same spatial limits and orientations are visible, and the same basic factors were at work. The pattern of spatial behaviour set by the Clann Mhuirich before 1705 was continued by them into the mid-nineteenth century; it was also the pattern followed by the other kin-groups in the community. Endogamous alliances in several kin-groups are oriented variously towards lower Badenoch and Brae-Lochaber; endogamy also accounts for the long-distance migration of two women to Inchnadamph in Assynt and to Perth. The behavioural link between kin-groups is also well demonstrated by the agnatic migration of a half-pay officer of a Clann Mhuirich lineage to Burgie House in the parish of Rafford near Forres and the consequent movement of new spouses of other kin-groups from Laggan to that vicinity in 1797, 1848 and 1852, suggesting contact between the parishes over a period of more than half a century.

      Finally, it should be emphasized once more that while social interaction can be adequately explained in terms of social factors such as kinship and marriage, these are not sufficient to explain the initiation and persistence of contact. The general ecology of the situation through time demands the presence of sustaining


political and economic factors. Direct evidence for these is difficult to document in time series, but Figure 9 hints at two of them in single instances: central Highland districts like Laggan were traditionally deficient in bread-crops, and had to establish contacts with lowland districts with surplus meal, while the same Highland districts possessed a super-abundance of summer grazing, more than sufficient for local transhumant practices.

Seasonal exchanges of cattle and annual transactions in meal probably account for persisting fields of social interaction. The long-distance transhumance of 1766, in fact, represented a still earlier version of the distinctive channel of social interaction between the parishes of Laggan and Rafford, thus exemplifying the fundamental connection between economic and social elements in the spatial behaviour of Highland communities over lengthy periods of time.

Department of Geography
Memorial University of Newfoundland, St John's


[1] Hon. Stuart Erskine (ed.), Lord Mar's Legacies, 1722-27. Scottish History Society 26 (1896) 219


[2] Most lineages of the Clann Mhuirich of Badenoch and the Castlelands of Inverness have used the surname Macpherson since the mid-fifteenth century; some have used other surnames. At the same time this surname had independent origins in Argyle and Skye, where it was used extensively by lineages unconnected with the Clann Mhuirich.
[3] A. G. Macpherson, An Old Highland genealogy and the evolution of a Scottish clan. Scottish Studies 10 (1966) 1-43; and, An old Highland parish register: survivals of clanship and social change in Laggan, Inverness-shire, 1775-1854. Scottish Studies 11 (1967) 149-192 and 12 (1968) 81-111
[4] W. Gillies, rev. Sileas na Ceapaich. Scottish Studies 18 (1974) 143
[5] Inventory of Macpherson of Cluny Papers, 1571-1860. Scottish Record Office, Edinburgh.
[6] H. Paton (ed.), The Mackintosh muniments, 1442-1820 (Edinburgh 1903)


Return to the AGM Menu Page