surnames and identities

This paper1, based on a survey of 314 Oslo residents, investigates the relationship between surname and identity. The aim was to find out whether themodern individual experiences his or her surname as a part of his or her identity, and what bond exists between surname and locale. Late modern society typically reveals a fragmentation of individuals from family background and place of origin. A hypothesized outcome of this separation, envisages a further breach between the individual and the area their surname denotes. If one’s surname is experienced as part of one’s identity, what then is the main reason for this? Are there in fact different experiences of identity based on some typology of names borne by individuals?


S O L V E I G W I K S T R Ø M
University of Oslo a b s t r a c t This paper 1 , based on a survey of 314 Oslo residents, investigates the relationship between surname and identity. The aim was to find out whether the modern individual experiences his or her surname as a part of his or her identity, and what bond exists between surname and locale. Late modern society typically reveals a fragmentation of individuals from family background and place of origin. A hypothesized outcome of this separation, envisages a further breach between the individual and the area their surname denotes. If one's surname is experienced as part of one's identity, what then is the main reason for this? Are there in fact different experiences of identity based on some typology of names borne by individuals?
[1] i n t r o d u c t i o n In Norway, approximately 70% of the population bears a surname which is originally the name of a farming area. Thus, these surnames represent certain connections with specific places. As a result of this, it has been assumed that individuals with a farm name as surname may have an affective relation to the place the surname refers to. 2 Norwegian onomastics has its roots in the patriot movement formed in the latter half of the 19th century (Helleland 1999), and farm names have been viewed as carriers of the nation's distinctive character and history. In 1926, the Norwegian onomatologist, Magnus Olsen, asserted that the oldest farm names used as surnames were considered the most prestigious amongst Norwegian family names, and furthermore that these names were a reflection of the individual's place in society (Olsen 1976(Olsen [1926). Since the eighties, the concepts of post-modernism and globalization have widely been interpolated as encapsulating contemporary society. In the phraseology of sociologist Anthony Giddens, modern social life is characterized by a separation of time and space; this condition of articulation of social relations across wide spans of time and space is seen in disembedding mechanisms which separate interaction from the peculiarities of locales, [1] The current paper is a re-edited version of my papers read at the two Names and Identities conferences at The University of Oslo (Wikstrøm 2007;Wikstrøm 2008) and further based on the main results of my Masters thesis (Wikstrøm 2009 [258] solveig wikstrøm and in institutional reflectivity which is the regularised use of knowledge over the circumstances of social life and forms a constitutive element in its organisation and transformation (Giddens 1991, 2). Modernity is essentially a post-traditional order, where the globalising tendencies are inherent, and thus, one would imagine that late modernity has caused a certain breach in the name-bearer's view that the surname is a means for showing background related to family and "homeland".
[2] s u r n a m e t y p e s i n n o r way Until the latter part of the 19th century, most Norwegians didn't carry surnames. The common name system provided the individual with a Christian name (in some cases more than one) and a patronymic whose function was to show who the bearer was son or daughter of. In addition, the individual could be characterized by the name of the farm where he or she lived or came from. This name could change if he or she changed dwelling place. When the surnames became fixed as result of the first Norwegian law on personal names of 1923, the farm names and patronymics thus became the distinctive types of Norwegian surnames (Nedrelid 2002, 134). Surnames in Norway have been classified, as per Helleland, into five main groups: (i) Surnames from place-names, mainly farm names (ii) Patronymics (ending with -sen) (iii) Surnames brought into the country some hundred years ago by officers of the crown or craftsmen (iv) Surnames from recent immigration (v) Newly designated, "ornamental" surnames; often of the same type as group (i) (Helleland 1997, 51).
Group (i) is, as mentioned above, the largest group. Some farm names are ancient, often more than a thousand years old, in both non-compound names such as Haug ('hill') and names compounded with for example -vin or -heim, where the ending is usually reduced, e.g. to -en and -um as in Vøyen and Holum. Often the more recent names refer to smaller farms and are given the definite form (e.g. Haugen, 'the hill'). These names are relatively frequent. 26 % of the population bears a hereditary patronymic in group (ii). One masculine compound with -sen ('son') is the most frequently occurring surname in Norway -Hansen. The rest of the population bear foreign or constructed names. Some foreign names have been used in Norway for some hundred years, first mainly by Danish, German and Scottish officers of the crown or craftsmen (iii), such as Meyer, Møller and Smith. surnames and identities [259] Even if these names are frequent in their countries of origin, they are however relatively infrequent in Norway and commonly associated with the bourgeoisie. Similarly, names from recent immigration (iv) are primarily a city phenomenon. Obviously, this is a very complex group, consisting of names from all parts of the world -the most frequent names from this group in Norway are Nguyen and Ali. The constructed names in group (v) are primarily of Swedish descent, as result of a tradition with ornamental names (Brylla 1995). Frequent names of this type are Lindberg and Berglund. Also belonging to this group are new names constructed on Norwegian ground which are aimed to resemble traditional farm names.
[3] q u e s t i o n n a i r e The survey consisted of 22 questions, 13 for those who didn't need to answer follow-up questions. Further, it was separated in 9 parts. In brief, the questionnaire asks about what the present surname is, former surname for those who have changed name, reasons for name change, geographic connection, etymology, like or dislike of the surname, and the extent to which the surname is a felt to be a part of the individual's identity. At the end of each part a text box is provided to encourage the respondent to give further comments. The text boxes are important as they potentially supply valuable knowledge in additional information which otherwise might not be covered by the declarations in the questionnaire. The themes in the questionnaire are thoroughly described in Wikstrøm (2007). […] the Web is evolving as a mode that must be used in conjunction with more traditional modes such as telephone and mail. This means that there is an increase in the occurrence of mixed-mode surveys, whereby some data are obtained by the Web and other data are gathered by another mode or modes (Dillman 2007, 450).
In light of this, I chose to have several groups of respondents who were invited through different channels (such as members of a sports club or visitors to a specific web page); in all, six different groups participated in the Web survey. A control group was contacted by telephone as a random sample and asked to participate by filling in a paper version of the questionnaire. A total of 314 answers was obtained, with 49 of these constituting the control group (Wikstrøm 2008). Surveys on the Web require access to a computer online and some skill using it. This means that the more resourceful are assumed to answer the survey. I tried to check the partiality by comparing the answers I got from the different groups to record any potentially substantial differences in answers. This comparison revealed that the six groups who had answered the questionnaire on the Web did [260] solveig wikstrøm not differentiate essentially from the control group. However, the groups differentiated from the population as a whole, according to figures from Statistics Norway (SSB), as overall the respondents proved to be slightly younger and considerably better educated. Thus, I have reached a specific part of the population. Unn Røyneland states that traditional sociolinguistics requires the selection of respondents to be representative of the actual population; for example, covering all the members of a particular language group or dialect (Røyneland 2005). However, following Røyneland, a specific group of the population can usefully be investigated in respect to attitudinal meanings. In the interpretation of Broady, sociologist Pierre Bourdieu and associates seldom use random samples of respondents in investigation. Rather, they are interested in the individual and groups, in their attitudes and qualities which are representative in the sense that they portray certain social phenomena of interest (Broady 1991). The group I wished to find out about was urban citizens, those who presumably have a distanced relationship to their often rural family names. The respondents are accordingly not representative for the population as a whole in statistical sense, but they are representative for the part of the population I wished to reach.
[5] c h a n g e o f s u r n a m e By asking if, and then why, the respondent has changed her surname, I wished to find out if there are certain surname types which seem to be preferred. The most recent Norwegian law on personal names, in effect from January 1, 2003, has enabled greater freedom to choose the name that best expresses one's identity. Still, there are no respondents in my survey who state that they changed their surname in direct consequence of this law. Most respondents changed their surnames as a result of marriage: 44,6 % of women and 4,8 % of men who are married changed their surnames. I consider the numbers too low to be represented usefully by percentages, but in general, for both sexes, the patronymics seem to be the name type which loses on the name change. Some respondents state that they experience these surnames as "too common". Furthermore, the women generally seem to change name as a result of tradition. Many, especially younger women, keep their maiden names as middle names, as an expression of their initial social and personal identity. Many children are given their mothers' maiden name as a middle name as well, with some women stating the reason for this is that the child's attachment to both branches of the family should be shown. The few men who claim to have changed surname seem to have changed to a name which is commonly associated with a higher social class. This is in accordance with the findings of sociologist Renee Thørnblad, who concludes that women gain respect when they follow the tradition and change their names to their spouses', while men gain respect when the new name in itself gives higher status (Thørnblad 2003). The names frequently preferred are those of type (i) surnames and identities [261] and (iii), as outlined in section [2] above. Other name changes show that the respondents use the freedom they have to choose among traditional farm names or create new names with a basis in the place-names they have a family relation to. The material shows no cases of change in spelling to achieve what traditionally have been regarded as status names, for example by change to older orthography (e.g from Kvam to Quam), or by deletion of the definite article (e.g from Haugen to Haug), or with the hyphenation of two family names (Berg-Hansen). The change of surname seems to incorporate a certain conservatism, whereby the relationship to one's family, continuity and, perhaps, as Olav Veka claims, fear of being considered pretentious (Veka 2001). The results show that no one has made name changes which can be said to signify a break with Norwegian naming traditions.
[6] s u r n a m e a n d p l a c e 61,8 % of the respondents have a surname from a Norwegian place-name, and this result is not inconsonant with the country as a whole. More have a foreign name of type (iii), (iv) and (v) shown in section [2] by comparison to the rest of the country, as is expected within urban populations. Many of the respondents are able to supply the geographical origins of their names, with 87,1 % knowing the county. However, only 52,1 % of the respondents with a farm name as a surname considered, upon direct questioning, that the connection to this place is meaningful. Even though this amounts to more than the half of the respondents, I claim that it is a relatively low number when we relate this to Norwegians' presumably strong connection to their "homeland". 27,3 % assert that the connection reflected in their name is not important. Thus, being able to trace the surname back to a specific place does not seem to give attachment to this place as a natural consequence. Three of those with surnames from place-names have surnames from farms in Oslo. Of the rest, 61,8 % have a physical connection to the place their surname refers to, as they have grown up there or their families came from that place. Compared with the proportion of respondents who reveal that the connection is regarded as meaningful, these numbers may show a certain disparity between what we can call an actual connection and the connection one experiences through the name. 12,9 % of the respondents with a farm name as a surname claim that their surname does not originate from a specific place. 10,3 % have not answered the question which concerns relation to the place. Thus, the respondent doesn't always have a notion that the surname she bears is (or has been) used for a specific place, and this obviously exposes a weak relation to this place. Questions concerning provenance are, of course, not so relevant for those with patronymic surnames. The other name types reveal that approximately a third of the respondents with names of type (iii), (iv), and (v) feel an attachment to the place or country their name comes from. However in group (iv), consisting of respondents with foreign names from recent immigration, not many claim such [262] solveig wikstrøm an attachment to place: 6 of 24 respondents. Thus, to understand the identity which is associated with the surnames of foreign origin, it is probably reasonable to put less focus on the assumption that the names say something of importance about the attachment to a specific place, as some of these names build on very different cultures and mentalities. For example, both respondents who claim that they or their family come from Pakistan, gave additional information in the text boxes explaining that their names are primarily Muslim and thus less associated with a certain country. Cases where the attachment to place via the surname is apparently of low import to the name-bearer, may suggest that the name has another function than tying the bearer to a place of origin. Whereas, for example, surnames from farm names have their parallels in toponyms, they actually function as anthroponyms; while some respondents don't recognize their surname as a toponym at all. Gudlaug Nedrelid states that most toponyms give meaning to Norwegian speakers, whereas anthroponyms are more frequently empty of meaning (Nedrelid 1993, 169). As mentioned above in section [2], surnames from farm names and the patronymics were in pre-modern society used as by-names, and consequently names which characterized the bearer by stating, for example, which farm she came from or who her father was. Today, these names are fixed surnames. Kristin Bakken suggests that when patronymics became fixed, they lost their original, specific, meaning, and the new meaning can be something like 'family connection' (Bakken 1995, 43). This may have occurred in respect of the original farm names as well, that is, they might have been emptied of their original meaning, for example 'place connection' and the new meaning may be 'family connection'. If so, we might say that the anthroponym has, to some degree, displaced the toponym.
[7] e t y m o l o g y The respondents were asked if they know what their surnames mean, and, overall, slightly under the half, 47,1 %, claim they do know the meaning. The highest proportion of these -but nevertheless surprisingly low -was the group with surnames derived from patronymics. Here, 56,9 % state that they know the meaning, and all elaborate that their names are compound with a man's name and 'son'. Of those with surnames from farm names, 46,9 % claim to know the meaning of the name. Most furnish this in the text boxes following the question. Names with an etymology in accordance with modern Norwegian, like Haug 'hill' and Sandvik 'sand' and 'bay' are relatively often understood by the respondent, as are more obscure names like ancient names compound with, for example, -vin. Yet, many respondents claim not to have investigated what their names mean (33,5 %), and some of these names have an etymology easily comprehensible to Norwegian speakers, such as Bakken 'the hill', Hauger 'hills' , Torp 'croft', and Wold 'meadow'. It is of course difficult to conclude which appellatives native Norwegians "should" surnames and identities [263] understand, but these examples may show that the surname to a certain extent has lost its connection to the language for the bearer. Both amongst those with surnames from farm names and those with patronymics, there are some respondents who state that their names don't have a meaning at all (4,6 % and 19,4 % respectively), and for the farm names concerned, this also includes names one might think should be possible to understand for native Norwegians (e.g. Huse 'house(s)' and Øien 'the island', 'flat area bordering water').
I will claim that the fact that many of those who don't understand their name, or parts of it, is not owing to their possessing names with more obscure meanings. Quite the reverse, many have knowledge about the etymology of the older names, and their answers often reveal that they represent a more educated part of the population. The division is thus not only between those with names which are "easy" or "difficult" to interpret, but also between those who experience that their names have a meaning and those who don't experience such a quality. This might show that some have a notion that their names have another, different, or "deeper" meaning -a meaning which is detached from the language. In section [6], we learned that some respondents didn't experience that the farm name they bear as surname is or has been a place-name, and in some cases, this might have eventuated due to the name's usage in relation to other content linked to the expression. In this case, the name now functions as a surname, and thus the anthroponym has displaced the toponym. A similar process could have occurred in the case of influence over linguistic meaning. As it is a popular topic, many will be able to explain what their names mean. Still, the question is in some cases not understood. This can be explained by referring to Bakken, who suggests that the distinctive feature of anthroponyms is that their connection to the vocabulary is, in many cases, broken, that the relationship between the name and the bearer often is arbitrary, and furthermore that anthroponyms more often than toponyms are regarded as empty of meaning (Bakken 2000). As we saw in section [6], the farm names may have been emptied of their original meaning and the new meaning of the surname can thus be 'family connection'.
[8] l i k i n g a n d d i s l i k i n g o f t h e s u r n a m e The respondents were asked to answer the question "Do you like your surname?" by picking one of four alternatives. This showed that 84,7 % liked their surname, 4,5 % answered no, 3,2 % didn't know and 7,6 % were indifferent. The type of names most frequently valued positively are those brought into the country some hundred years ago by officers of the crown or craftsmen (group (iii) in section [2] above), at 93,3 %, followed by the surnames from farm names (i) at 91,8 %. The patronymics (ii) are the most disliked, but nevertheless, 68,1 % claim they like their patronymic, while 18,1 % are indifferent. Surprisingly many with names from recent immigration (iv) state they are indifferent, 20,8 %. Those who claimed [264] solveig wikstrøm to like their surnames were asked to pick among maximum three of twelve statements to describe why. The total percentage will thus exceed 100 %, and for this reason I will not detail the proportions. Family connection was most important value here for all groups except group (iii). All the respondents from this group have names which are uncommon in Norway, and they have chosen "The name is uncommon" as the most important reason. As a matter of fact, it is an important reason for all respondents who don't have one of the more common names. The second most popular reason is "The surname is "me"", as an expression of individual identity. These two, most popular, explanations -that the name shows family connection and that the name is "me" -stand in a dialectical relation, as they are expressions of social and individual identity. According to Brit Maehlum, the two types of identity cannot be seen as independent of each other, as the individual's identity is crystallized in the tension field between the individuality and the commune (Maehlum 2008). The formal properties of surnames, such as spelling, etymology, sound, and so on, do not seem to be of great importance when it comes to liking of one's name. It is more significant that the name is able to make the individual stand out from the masses in combination with showing family connection. In the comments which are given in the text boxes, many also tell that the first name in combination with the surname is important in order to stand out and be unique. On the other hand, the perception that the name is too common is the most frequently attributed cause for respondents who don't like their names, these most often are among the patronymic group (ii). As we saw, many in this group are indifferent to their names as well, and some experience them as dull. The surname is, though, a part of them after all -many claim that they would not consider a change of name in such cases. Of the entire survey, only two respondents express that they don't like their surnames owing to their experience of the name's negative connotations, with an obscure meaning in modern Norwegian and resemblance to Norwegian or English slang or swear words. In section [1] above, it was suggested that the oldest farm names, with their formal features, would be considered as status names, able to show the individual's place in the society. In light of this, we would expect a discernment between the older farm names and the younger names used to refer to smaller farms and cotter's farms. However, this distinction doesn't appear to be evident today, and is expressed neither by respondents who would gain nor those who would lose on the basis of such comparison. Most positive attributions given are not associated with class, for example respondents of group (ii) frequently experience their names as unpretentious, easy to spell, and easy to use abroad, which are stressed as good qualities. This may show that the individual focuses on and appreciates the typical features associated with the name in order to identify with it. The case where status is not greatly emphasized is a reflection of the names not revealing very much of interest about the bearer's background. In addition to the fact that most surnames and identities [265] city dwellers carry all sorts of farm names and patronymics as surnames, the urban community is complex, and we do not have the same tools today to classify somebody socially by their surname. It is thus possible that the exclusiveness of having an uncommon surname today has replaced the exclusiveness one had earlier by having a name type which implied high social class. The assertion just given, that having an uncommon surname is important for modern city dwellers, is emphasized by the assumption that one of the typical constructs of identity in the late modern society is the wish to be unique. The religious historian, Otto Krogseth, employs three criteria to comprehend modern identity: continuity, the feeling of being integrated, and individuality (Krogseth 2007, 60). The three most common attributions the respondents gave when describing why they like their names run concurrent with these criteria. 'Continuity' can be considered in relation to the link established by surnames to family. Some respondents also mention that the name has "become" themselves, and that they feel continuity by having a name they are used to. The feeling of being 'integrated' is described by Krogseth as our feeling of being able to carry all the diverse roles modern life charges us with. In that context, we can suppose that the surname collates and expresses who we are -the name is nevertheless "me" in a complex society. 'Individuality' is most cogently expressed for individuals having an uncommon name. According to Krogseth, we ask "Who am I, as distinct from others?", and he claims that individuality is confirmed in spite of, and often as a protest against, the threat of unification. This is typical for the post-modern formation of identity. Wanting to show individuality through having a name the individual shares with as few as possible could thus have supplanted the earlier exclusivity of having a name which shows social status -in a society where fewer master the codes for working out what names these are. In the text boxes, many respondents tie the fact that the name is "oneself" to the fact that it is uncommon and gives a feeling of uniqueness in the same reasoning. The name's link to family is not discussed by any of the respondents, and thus it seems that this is viewed as a matter of course. On the other hand, personal identity is often defined, and this is often clarified as evoking a feeling of uniqueness. As we have seen, this is most important for the group who in all cases have an uncommon name. When the respondents so clearly express that their reasons to like their names have parallels in the criteria for identity, this shows that to like one's name and relate one's identity to the name is, in many cases, two sides of the same coin.
[9] s u r n a m e s a n d i d e n t i t i e s Identity is a complex concept with so broad a range of associations that it must be characterized as a vague conception. As mentioned in section [8], the individual's identity is crystallized in the tension field between the individuality and the commune. According to Maehlum, there has been a change in the understanding of the notion of identity from representing a permanent, eternal and unchangeable condition, to be a condition subject to change and negotiation -named as essentialistic and constructivistic notions on identity respectively (Maehlum 2008, 109). As we have seen, it seems that feeling identity with one's surname is closely attached to liking it. In the society we live in, we wish to stand out, to show our own identity by expressing something genuine and particular which can not be reduced to something common. About as many who like their name simultaneously express a personal association of identity with their name (85,7 %). The groups who like their names best, as shown in the previous section, are coincident to the groups who feel strongest identity with their names. Moreover, those with uncommon surnames have the strongest identification with their names. Psychologist Kenneth L. Dion has worked on the relation between anthroponyms and identities. Even though his studies concentrate on first names, the results are comparable with the current analysis which reveals that a positive feeling towards one's name is closely tied to feeling identity with it. Dion even connects this with liking oneself: Identity and self-acceptance are different, though related, psychological concepts. Identity refers to a person's sense of who or what she/he is. Self-acceptance or self-esteem concerns an individual's evaluation or overall liking-disliking of themselves. It seems reasonable to assume that one's sense of identity and the extent of one's self-acceptance would tend to co-vary with one another. Similarly, from the idea of a close tie between one's name and one's self, selfacceptance and liking for one's own name would be expected to correlate positively with one another, such that greater liking for one's own name is associated with stronger acceptance of self.
[…] Individuals who liked their own name tended to identify more strongly with it, in the sense of affirmatively answering the question "Do you feel your first name is you?" Furthermore, individuals who either liked their first names or were neutral toward them scored higher on the self-acceptance scale than those who expressed dislike of their first name (Dion 1983, 251).
To find the most central cause in consideration of how a surname invests us with a feeling of identity, the respondents were asked to select one of five answers. The percentages are shown after the alternatives: (i) It gives a picture of who I am -14,9 % (ii) It connects me with my close family or in-laws -25,7 % (iii) It connects me with family history -44,6 % surnames and identities [267] (iv) It connects me with the place where the name comes from -9,2 % (v) Other reasons -5,6 % The two most frequent selections above reveal that attachment to family history and close family are highly significant. In this context, an aspect of philosopher Paul Connerton's theories is cogent. Connerton suggests that social memory is used together with historical reconstruction in order to form an identity based on family history. Here, 'social memory' is understood as the oral tradition concerning family history. Historical reconstruction is gained by using written sources to fill in and adjust social memory, and these sources are often understood as authoritative. Thus, they are able to form the memory of a group, in order to establish a fixed identity based on family and roots. This is, however, characteristic of modernity, which implies that both modern identity and family history can be negotiated and changed (Connerton 1989). In a historical reconstruction of an individual's family history, the individual can thus select and focus on what she feels best expresses her distinctive character. As we can see above, family history is weighted most heavily among the respondents; more important than the more proximate physical (in time-space) and psychological relations with the individual's contemporary family and in-laws. A construction of family history thus makes the individual able to accomplish the project of modernity and construct an identity which is autonomous and has broken free from the closer, contemporaneous, family circle. According to Thomas Hylland Eriksen, the post-traditional realization of identity leads to estrangement and autonomy, but the approximation of a family connection can be important in order to be able to take part in the society's collective memory. However, this is rooted in a deliberate choice where the aim is to show one's distinctive character in this society Eriksen (2008). The third most selected answer, "It gives a picture of who I am", is concerned solely with the individual's identity. Here, connection is of less importance. This can be said to represent a certain autonomy from the family connection the name represents. We can examine this in relation to Pountain and Robins' notion of "Cool", used to describe one of the mechanisms in modern society, where withdrawal from the traditional community is a distinctive feature: Cool [...] is a form of self-worth that is validated primarily by the way of your personality, appearance and attitude are adjudged by your own peers.
[…] Cool has moved from being the pose of a tiny minority and is fast becoming the majority attitude among young people (Pountain & Robins 2000, 153, 167).
The group who most frequently selected answer (i) above consisted of respondents with names brought into the country some hundred years ago from officers of the crown or craftsmen (group (iii) in section [2] above), and as we learned in [268] solveig wikstrøm section [8], this same group attributed the uncommonness of their names as the cause of their positive association to them. The notion of self-identity can be tied to the feeling of uniqueness through Dion's clarification of other scholars' theories on names as uniqueness attributes: […] the personality theorist Gordon Allport (1961) contended that one's name is the focal point upon which self-identity is organized over the course of an individual's life.
[…] C.R. Snyder and Howard Fromkin (1980) proposed that names, along with commodities as well as attitudes and beliefs, are "uniqueness attributes" through which individuals may differentiate themselves from other persons.
[…] our names are salient to ourselves and others, especially when they are relatively infrequent and therefore distinctive in the context of groups of which we are members (Dion 1983, 246, 249).
Names are thus salient for ourselves and the community around us, especially when they are less frequent. The theories Dion refers to, illuminate what represents the core of my investigation; that an uncommon name is best-liked and the best marker of individual identity. As we have seen, the surname's ability to show where the respondent has her roots seems to be of less importance. The option of, "It connects me with the place my name comes from", was selected by a mere 9,2 % of respondents.
Of those with surnames from farm names, group (i), 11,3 % chose this alternative. Aside from "other reasons", this is the least often selected alternative for this group also.
Therefore, it appears that the connection to place is generally of lesser importance, as reflected by the results of the present investigation. This suggests that a close association between surname and place, especially among surnames from farm names, is uncertain. Agnete Wiborg has investigated the connection between place and identity, and she claims that the tie to a specific place in current society departs from the traditional assumption of a close attachment between place and identity. Connection to place must be viewed as a tool in construction of the individual identity (my translation): Place of origin is more frequently disconnected from family, social class, and way of living, even if these connections are operating on a symbolic level and regulating how place of origin is experienced and given meaning. The assumption that there is a close attachment between individuals and places which is grounded in Norwegian ideology and way of thinking still exists, but has now come to a discursive level where these assumptions may be contradicted and negotiated surnames and identities [269] […]. The rural district or the place as a social, cultural, and geographical background represents a repertoire of symbols which individuals use in different ways to build their individual identity of choice (Wiborg 2003, 148, 149).
As we can see, place of origin is a difficult conception. When the individual doesn't live at the place from where her background originates, the attachment will be more marked by symbolic and emotive associations with place, rather than with place as constituted by 'real' interactions. Wiborg suggests that attachment to place is experienced symbolically, evocative of good memories and relations. When place gets detached in time and space, one can reconstruct it with forgetting or repressing what one does not choose to remember. For example, the scenery of the place can symbolize what the individual wants to remember from the past and the picture one has constructed of the place from the current position (Wiborg 2003, 137, 141). In light of this, and in light of the results presented in section [6] and [8], it seems that the place the surname originally connoted is subordinated. The surname is a bearer of one's identity because it is a means utilized in the individual's intentional construction of self, not because it attaches the individual to a "homeland". "Other reasons" respondents gave clarifying why the surname is experienced as an important part of one's identity can be divided into two streams of responses: The name's ability to make the bearer unique and the name's ability to ensure continuity. These are both -as we saw in section [8] -important aspects of why the individual likes her surname.
[10] c o n c l u s i o n What I have tried to portray situates the complex of modernity whereby national and local traditions meet global trends, as illustrated through a group of city dwellers' relationship to their own surnames and their geographic origin. Globalization and late modernity have influenced many individuals' attachment to family and homeland, and I would suggest that the distinctive characteristics of global society have marked many individuals' relationship to their own surnames as well. What seems to be important is that the surname makes it possible to point out the individual as unique in her community, and this community isn't necessarily tied up to the notion of "homeland" expressed through the surname. As we have seen, Giddens (1991) proposes that human relations are separated from time and space in late modern society. In light of this, one might say that many individuals seem to be marked by what Zygmunt Bauman (Bauman 2001, 56) summarizes as the message of the 'cosmopolitan' way of being: "it does not matter where we are, what matters is that we are there".