(Man (Journal of the Royal Anthropological Institute), 1994, 29(4))

Gísli Pálsson

University of Iceland


By focusing on Icelandic fishing, this article suggests that personal enskilment, in both fishing and doing ethnography, means not to mechanistically internalise a stock of knowledge but to be actively engaged with an environment. Such a perspective - informed by theories of practice - resonates with certain aspects of Icelandic discourse; when discussing enskilment, Icelanders sometimes refer to learning about fishing as the recovery from seasickness, 'getting one's sea legs' (að sjóast). Practice theory, I suggest, has important implications for studies of economic production and differential success, in particular discussions of the 'skipper effect'. Skills are bodily dispositions with differential distributions, but to isolate their acquisition and application from everything beyond the boundaries of the soma would be to subscribe to a reductionist theory of both learning and sociality. The alternative view of practice theory emphasises the importance of attending to whole persons, master-apprentice relations, and the wider community to which they belong - decentering the analysis of enskilment and craftsmanship.


To take a decentered view of master-apprentice relations leads to an understanding that mastery resides not in the master but in the organization of the community of practice of which the master is part . . .

Lave & Wenger 1991


When talking about seasickness, Icelanders implicitly recognise the relationship between knowledge and practice, and the unity of emotion and cognition, body and mind. For them, 'seasickness' (sjóveiki) not only recalls the bodily state of nausea sometimes caused by lack of practical knowledge, the unexpected rocking movements of the world, but it is also used as a metaphor for learning in the company of others, which is seen in terms of the recovery from seasickness - 'getting one's sea legs' (sjóast). Those who return 'home' after an experience on rough seas are said to have found their sea legs (verða sjóaðir); as nausea is replaced by well-being, a quantum leap in learning and sociality takes place. In this article I shall argue that getting one's sea legs - becoming skilful - means to attend to the task at hand, actively engaged with a social and natural environment. This suggests a notion of enskilment that emphasises immersion in the practical world, being caught up in the incessant flow of everyday life - and not simply, as many cognitive studies have assumed, the mechanistic internalisation and application of a mental script, a stock of knowledge or a 'cultural model', 'what one needs to know in order to behave as a functioning member of one's society' (Holland and Quinn 1987: 4).1

The approach adopted here is informed by theories of practice and practical knowledge recently advanced in several disciplines, including anthropology (Bourdieu 1990; Lave 1988; 1990; Ingold 1993a; 1993b), psychology (Resnick 1987; Lave and Wenger 1991; Reed 1993), sociology (Joas 1993), and economics (Marglin 1990). Such an approach was outlined many years ago in both European and North American contexts - by Malinowski (see Leach 1970), Veblen (1964[1914]), Polanyi (1958) and some others (Joas 1993); only recently, however, has it begun to emerge as a consistent theme with a momentum of its own (Ortner 1984), reinforced by current theorising on the body (see Lock 1993). The perspective of practice theory, I suggest, has important implications for studies of differential fishing success, a theme explored by a number of anthropologists (see, for instance, Barth 1966; Acheson 1977; Löfgren 1977; 1989; Byron 1980; 1986; Kottak 1992), especially discussions of the so-called 'skipper effect', the idea that the contribution of the skipper is critical to the size of his catch relative to that of others (Pálsson and Durrenberger 1982; 1983; 1990; 1992; Durrenberger and Pálsson 1983; Gatewood 1984; Acheson 1988; Thorlindsson 1988; White 1989; McGoodwin 1990; Gatewood and Mace 1990; Bjarnason and Thorlindsson 1993; Durrenberger 1993). Discussions of differential fishing success, I argue, often remain couched in reductionist terms, emphasising the notion of the autonomous individual, sometimes at the expense of both ethnography and theoretical insight. Acheson observes that the 'technical skills' of the fishing captain 'are far more critical in determining the catch of a boat than any other single factor' (1977: 111, emphasis added) and much research has attempted to establish to what extent that is so. The problem, however, is that it is far from clear what exactly the reference to 'technical skills' as a 'single factor' should be taken to mean.

Readdressing the issue of fishing skills, and emphasising the social nature of human action, inevitably shifts the ethnography to new kinds of questions. Focusing on everyday activity and situated practice rather than individualistic, normative models - as Lave points out (1988: 15) in another context - 'motivates . . . a different set of problems and questions than the study of virtuoso performance and peoples' failures to produce such performances'. In particular, it encourages us to attend to whole persons, master-apprentice relations, and the wider community to which they belong, decentering the study of enskilment. If enskilment is a necessarily collective enterprise - involving whole persons, social relations, and communities of practice - so also is ethnographic apprenticeship, doing anthropological fieldwork (Gudeman and Rivera 1990; 1993). I will argue, drawing upon my experience in my native Iceland, that enskilment in fieldwork inevitably involves psycho-somatic processes, if not veritable 'gut reactions'. Fieldworkers usually begin their 'trip' on the margin of the community, nauseated by their novice status. As they become increasingly involved in and knowledgeable about the activities of others, they move towards the centre and begin to feel 'at home', in both their bodies and the company of others.


Learning theory

Theories of learning and craftsmanship have often been presented in highly normative terms, presupposing a natural novice who gradually becomes a member of society by assimilating its superorganic heritage (see Bloch 1991: 189). Although different social theories have different notions of the relationship between the individual and the collective and of what sociality entails - thus Boasian culture theory speaks of the 'enculturation' of individuals as cultural knowledge is stored in their heads, while for Durkheimians 'collective representations' are supra-individual byproducts of interaction - in normative theory learning is generally assumed to take place with socialisation, broadly defined as the acquisition of a stock of knowledge about expected ways of thinking, feeling, and acting (see Ochs 1988: 5).2 Given the normative approach, learning entails the transmission of culture, a mental code or script that exists prior to and independent of human activities, a recipe for action (a prescription) analogous to a book of grammar or a dictionary - in short, a Saussurean langue. Related to the notion of learning as cultural transmission is the tendency to think of the person in terms of a container, 'as if the person possessed a fixed capacity [of competence] analogous to the amount of liquid that can be placed in a glass' (Fischer, Rotenberg, Bullock, and Raya 1993: 94).

There are several serious problems with such an approach. For one thing, in reducing the novice to an imitator of technique, the normative approach leaves unanswered the fundamental question as to how a body of cultural knowledge is constructed. If both novices and tutors (by definition, former novices) are mere recipients of models and texts, submitted to a body of knowledge given in advance, it is difficult to see how human creativity could be possible at all. This problem is evident in much of the literature on 'indigenous knowledge' and 'ethnoecology'; as Ellen remarks with respect to this literature, people 'just do not conduct their lives in . . . pre-programmed ways' (1982: 225). Also, the model of cultural transmission is highly ethnocentric - the product of western history and textual discourse, reinforced by the tradition of literacy and the institutions of formal schooling and disembedded training - caught up in dualisms of mind vs. body and learning vs. context (Suchman 1987; Resnick 1987). Western tradition is preoccupied with analytic and theoretical ways of knowing, episteme, devaluing and misrepresenting practical and contextual knowledge, techne (Marglin 1990).3 This tradition needs to be explained and not taken for granted in a truly comparative study of epistemology.

The normative view, moreover, misconstrues the essence of the lived-in world, failing to capture what it means to engage in a skilful act, the 'feel' for the game. Rosaldo rightly remarks, using the game of basketball as an example, that ethnographic narratives of complex events must go beyond descriptions of rules and strategies; 'there is a vast difference', he points out, 'between merely knowing the rules of the game and having the knowledge required to follow the game in the manner of an experienced fan of fine judgement' (1980: 22). What is most important, however, is the construction of a third kind of knowledge, conveying the essence of the game from the point of view of the player. This is precisely the approach Rosaldo seems later to have adopted, in his critique of the objectivist ethnographer who 'positions himself as a spectator who looks on from the outside' (1987: 99). While rules and representations are relevant for the participant - the player - they are not what learning is all about. As Ingold argues, the novice becomes skilled not through the acquisition of rules and representations, 'but at the point where he or she is able to dispense with them' (Ingold 1993b: 462):


rules and representations . . . are like a map of an unfamiliar territory, which can be discarded once you have learned to attend to features of the landscape, and can place yourself in relation to them. The map can be a help in the beginning to know the country, but the aim is to learn the country, not the map.


Indeed, a person who cannot use his or her knowledge to do something is 'like a man who collects maps but never takes a trip' (Miller, Galanter, and Pribram 1960: 2, cited in Gatewood 1985: 199).

Finally, normative theory assumes a one-way, hierarchical ordering of knowledge, as Lave points out (1988: 8):


In this theory, duality of the person translates into a division of (intellectual) labor between academics and "the rest" that puts primitive, lower class, (school) children's, female, and everyday thought in a single structural position vis-à-vis rational scientific thought.


Normative theory, then, necessarily dismisses everyday language as cheap and theoretically irrelevant, as 'mere' household words or 'loose talk'.4 Oddly enough, anthropologists, who at the beginning of fieldwork normally assume the role of the apprentice, sometimes shift to the heroic role of the privileged 'observer' at a later stage in their work; when in the field, in the process of writing down, they present themselves as naïve novices and their 'informants' as their tutors, but as soon as they return to their academic bases the roles are sometimes mysteriously reversed, as if the act of writing up necessarily presupposed a kind of communicative metamorphosis. The humble novice anthropologist becomes a tutor. The textual 'ups' and 'downs' of ethnographic research are separated not only by physical movement; the journey 'back home' represents a rite of passage during which the anthropologist supposedly experiences an instant coming of age - and meanwhile the native is trapped in the mythical 'ethnographic present'. Malinowski, as Leach points out (1970 :134-5), ridiculed the validity of native accounts - even though they might turn out to be close to or identical to his own.5 An example which has a bearing on the ethnography of Iceland is the recent work of Hastrup (1992, 1993), significantly written partly in response to the critique of an indigenous anthropologist (see Einarsson 1990). Hastrup attempts, emphasising the 'violence' and 'hierarchy' inherent in the anthropological project (1992: 122), to advance an argument which would allow her to ignore any argument developed by the 'natives', anthropologists and non-anthropologists alike.6

Practice theory offers an alternative view of learning, craftsmanship, and ethnography - a view that allows for a novice whose sociality is given right from the beginning. Assuming a social or constitutive model of the individual is to introduce purpose, agency, and dialogue into the process of enskilment - a radical break with the Cartesian tradition of separating ideas and the real world, learning and doing, experts and laypersons, knowledge and practice. The theory of practice, then, suggests a change in gestalt which 'shifts the boundaries of activity well outside the skull and beyond the hypothetical economic actor, to persons engaged with the world for a variety of "reasons" . . .' (Lave 1988: 17-18); the proper unit of analysis is no longer the autonomous individual separated from the social world by the surface of the body, a natural being who passively internalises the mental scripts of the cultural environment, but rather the whole person in action, acting within the contexts of that activity.

Such an approach - informed by practice theory and the notions of situated action, mutual enskilment, and communities of practice - emphasises both democratic communion and the continuity of the social world. While the relations of novices and their tutors are rarely those of equals, there is quite a difference between the open and linear system of apprenticeship and the dualistic structure suggested by normative learning theory. The former mode, as Reed remarks, invoking the 'ecological' language of J.J. Gibson (1979), involves joint intentions and collaborative instruction, 'in the sense of helping others to learn the affordances of objects and tools within the context of a given skill' (1993: 54), whereas the latter involves a pyramidal context and a formal method of teaching.


'Getting one's sea legs'

Moving from learning theory to ethnographic practice, my own Icelandic fieldwork (mainly in Sandgerði on the south-west coast of Iceland) provides a useful starting point. In the early phases of my fieldwork, the process of establishing the necessary trust and rapport often seemed painfully slow. At one point, however, in the middle of the winter season in 1981, I felt as if I made certain and sudden progress. A skipper invited me to join him on one of his fishing trips: 'If you really want to know what the fishing industry is all about, you must go fishing'. I accepted his offer, and shortly after we left harbour we were in rough seas. While the skipper and his crew patiently waited for the weather to improve, however, I was busy emptying the contents of my stomach. Then, miraculously, it seemed, the weather improved and, fortunately, the feeling of seasickness vanished. Quite suddenly, my nausea had been replaced with a sensation of alertness and well-being, an 'oceanic' feeling. I had, quite literally, 'found my sea legs'.

Many other fieldworkers have referred to similar 'breakthroughs' - signals of dramatic, positive change - in their relationships with the people they visit (Okely 1992: 17); such stories are often used in modern fieldwork rhetoric. For me, however, the metaphoric association with the journey draws attention not so much to privileged excursions into the exotic as to the similarities in doing fieldwork and fishing in rough seas. Seasickness, in fact, has much in common with the psycho-somatic symptoms identified, by both fieldworkers and medical experts, as 'culture shock' (see Wengle 1988: 9-10). It would be wrong, of course, to simply dismiss the experiences of seasickness and culture shock as pathological ones; both involve bodily manifestations of a particular stage in learning. The fishing trip, then, served as a useful reminder of a critical stage in my fieldwork, and more generally, the emotional and physical manifestations of mastery and enskilment. As many had before me, I came to realise that to engage in anthropology is not just to 'observe' and record but to participate in the lives of other people.

While I find it useful to refer to my fieldwork as a fishing trip, I do not regard it as a solitary trip across cultural boundaries. During my original fieldwork, of course, I relied heavily on conversations with local friends and key informants, and later on I collaborated extensively with another anthropologist (Paul Durrenberger). Whether we liked it or not, we were rarely able to enjoy the luxury of the detached ethnographer. Anything we had to say on the subject of common interest - differential fishing success - inevitably got entangled in ongoing debates among indigenous fishermen and boat-owners on the role and nature of individual contributions, on fishing expertise and resource management. We came to the conclusion (Pálsson and Durrenberger 1992) that anthropology was best described as a dialogic enterprise involving both feedback and disagreement, as a Malinowskian 'long conversation' (Gudeman and Rivera 1990) in which the acts of 'writing down' and 'writing up' were difficult to separate. This recognition - triggered by discussions with fishermen and social scientists and the reading of anthropological texts - sent me back to my early fieldnotes, to the communities where I had worked, and to the comparative ethnography. Somewhat ironically, it was only some years after my experience of seasickness, fishing, and collaborative ethnography that I began to suspect that the perspective I had adopted with respect to my own enskilment as fieldworker and ethnographer might be applied to the context I had studied - to fish finding and travels at sea.

In fact, the contrast made above, between the textual approach of normative theory and the situated approach of practice theory, may well be illustrated with reference to the ethnography of fishing and ocean navigation. An article by Gladwin (1964), which compares the strategies used by European navigators and by Trukese sailors when travelling from one island to another over miles of empty ocean, contains some pertinent observations. In the case of Europeans, Gladwin suggests, most of the necessary thinking is typically done in advance of its implementation, whereas in the case of the Trukese, decision-making is continuous, accommodated to winds, tides, ocean currents, and so on. Also, while European navigators are able to describe in words the procedures they follow and are likely to give a 'logical' explanation of what they are doing, a Trukese navigator 'cannot possibly put into words all of the myriad perceptions which have led him to be sure at that moment where the island is' (Gladwin 1964: 174). The European navigator operates deductively, proceeding from general principles to details, whereas the Trukese navigator seems to operate inductively; 'He does indeed start with details, but he never arrives at any discernible principles' (1964: 175). Gladwin's description, then, captures some of the differences between the abstract, textual logics of much of European navigation and the situated, intuitive procedures of Trukese navigators (see Hutchins 1983; Suchman 1987: vii-x; Resnick 1987). His analysis, however, tends to be phrased in the normative terms of western educational discourse, emphasising mental models and 'cognitive strategies' - what goes on inside the brain: 'The Trukese navigator', he argues, 'does it all in his head. This is an astounding intellectual achievement' (1964: 172). Gladwin is quite right in pointing out that Trukese navigation is a remarkable achievement - we may refer to 'this kind of ability as a "knack"', he says, 'and respect a person for his competence' (1964: 175). Anxious to show, however, that Trukese are able to travel long distances on open boats without the use of the 'external' navigational devices of the European, he accepts uncritically the mentalistic premises of the normative theory of craftsmanship.

In an article on salmon fishing in Alaska, Gatewood (1983) similarly emphasises intuitive, situated procedures, contrasting them to 'rational' procedures of decision-making by means of terms not unlike those used by Gladwin. For Gatewood, a 'rational' procedure involves conscious consideration of alternatives, deliberation in terms of these considerations, thoughtful reflection on the process itself, and adherence to well-define procedures for selecting the right alternative (Gatewood 1983: 348-9). An intuitive or 'reasonable' procedure, he argues, is the same as a rational one 'except that the manner by which the final synthesis is accomplished cannot be specified by the decider in advance of the decision itself' (1983: 348-9). In Alaskan fishing, the crew expect their skipper to make a decision on a 'rational' basis - 'as opposed to basing it on hunches, sheer randomness, dreams, divination, etc.' (1983: 348), but contrary to the image of rationality they try to protect, under severe constraints of competition and impression management, 'most skippers fall short of explicit, algorithmic procedures when it comes time to decide' (Gatewood 1983: 349). Gatewood's explanation for the discrepancy between image and reality is not simply that rational solutions to the decision dilemmas of fishing are unavailable; more importantly, skippers' decisions are 'a matter of unspecifiable skill, of tacit knowledge' (Gatewood 1983: 363). Shifting the focus of enquiry 'from normative and outcome-predictive to ethnographic', Gatewood argues, 'reveals how people differ from the machines and mathematical models they create' (1983: 364).7 Bloch similarly points out (1991: 190) that if we think of decision-making as a normative serial process carried on by a single processor, everyday tasks which a skilled practitioner adequately deals with in only a few seconds inevitably become Herculean in size. Bloch's perspective of 'connectionism', which postulates a different 'cognitive mechanism' (multiple parallel processors) fails, however, to account for bodily knowledge and communities of practice, emphasising instead knowledge as existing in the brain.

The perspective of intuitive knowledge and situated practice represents a radical challenge to cognitive anthropology and its preoccupation with intellectualised static systems. Few people actually operate in everyday life in the idealised manner described for the European navigator and the 'rational' Alaskan skipper, and those who do are unlikely to do so on a regular basis. Many people act, much of the time, on the basis of intuitive knowledge and bodily dispositions - 'tacit' knowledge (Polanyi 1958) or techne (Marglin 1990). As Suchman suggests (1987: ix), 'plans are best viewed as a weak resource for what is primarily ad hoc activity. It is only when we are pressed to account for the rationality of our actions, given the biases of European culture, that we invoke the guidance of a plan'. To adequately represent situated practice, the flow of unfolding activity, and how we find our sea legs, other models are needed. And this brings me back to differential fishing success, Icelandic fishing, and the notion of the skipper effect.


Differential fishing success

In fisheries generally, catches vary from one boat to another. As Sahlins observes (1972: 73-4), 'for certain forms of production, notably hunting and fishing, the likelihood of differential success is known to common sense and experience'.8 The nature and causes of such differences, however, are matters of some disagreement among anthropologists. In the literature on fishing, there has been a tendency to emphasise the variability of the skills of the leaders of fishing operations. Reviewing the literature, Acheson points out (1981: 290) that 'the vast majority of anthropologists are convinced that . . . differential success is primarily due to marked differences in fishing skill'. Thus Byron notes for the Shetland that how 'successful the crew is in economic terms depends almost entirely on the skipper's "nose" - his ability to make correct guesses about where to find fish' (Byron 1980: 228). At the same time, several ethnographies have drawn attention to the social world to which the skipper belongs. Some have focused on the importance of the crew for skippers' decision-making and, by extension, fishing success - a classic example being Barth's analysis (1966) of the role of trust and self-confidence in Norwegian herring fishing (see also Heath 1976; Johnson 1979; Bjarnason and Thorlindsson 1993). Some accounts, too, have emphasised that the world of the ship should not be presented as an 'endogenous' one, as a 'closed cultural system' (Byron 1986: 96). In the latter view, any study of what happens on board fishing vessels must consider the role of fleet behaviour (Durrenberger and Pálsson 1986; White 1989) and the larger contexts of fisheries and fishing communities (Löfgren 1977; 1989; Durrenberger and Pálsson 1987; Levieil and Orlove 1990; Kottak 1992). To what extent, then, does the focusing on the personality and production value of the fishing skipper contradict the emphasis on communities of practice (crews, fleets, and fisheries) and the situated nature of learning and decision-making?

When explaning differential fishing success, Icelanders have often assumed, during most of the twentieth century, that the contribution of the leader of fishing operations overshadows that of everything else. Catches are said to vary from one boat to another during any one fishing trip, a single season or a generation largely because skippers are different. In competitive fishing, the skipper's position or 'seat' (sæti) in the local hierarchy of catches is a central concern; if the skipper has a low position, he risks losing his job. Once during my fieldwork, a skipper was fired at mid-season by his company because 'he did not fish enough'. Being at the head of the fleet for the season, on the other hand, being aflakóngur (literally, 'catch-king'), brings a high degree of honour and prestige. One of the central points in Icelandic theories of fishing success, then, is the attribution of particular qualities to the 'good' skipper which enable him to catch more fish than others. The chief quality of a good skipper is held to be a very personal one, perhaps comparable to 'intelligence' in psychological testing, a quality that is independent of the frequency of its application. For Icelanders such a distinction is just as important as the distinction made by many educators between the 'cleverness' of a student and how often he or she puts it to work, the difference between intelligence quotient and diligence. The dominant Icelandic theory of fishing success, then, is hierarchical in that it emphasises differences among a group of producers, normally within a local fleet.9 As I shall argue later on, however, another (and somewhat less visible) element of indigenous discourse also draws attention to the co-operation of the crew and the wider context of communities and fisheries.

In Iceland, the hierarchical model of fishing success first developed at the beginning of the twentieth century, when the domestic economy gave way to entrepreneurial fishing and large-scale capitalist production for an expanding foreign market. In order to appropriate fish, skippers had to become fishers of men. Recently - especially after the introduction of a quota system in the cod fishery in 1983, a few years after my original fieldwork - a new discourse has been developed in competition with the earlier one. The present system of quota management allocates a given share of the annual catch, a transferable quota, to each boat-owner. Fishermen and managers agree that, as a result, the 'hunting element' of fishing (veiðimennska) is rapidly disappearing. Sometimes they express the view that, with increasing governmental control of the industry, the custom of awarding the most successful skipper of the year a particular prize on Fishermen's Day (a standard custom in most Icelandic fishing communities for decades) is a little archaic; as fishing is being 'reduced' to business transactions, success becomes less a matter of fishing skills than capital and economics. The top skippers are 'simply privileged quota-kings (kvótakóngar)'. A similar discursive change has been described for some other contexts where a quota system has been instituted; see, for example, Miller and van Maanen (1979).

Keeping in mind the importance of the hierarchical model in both the ethnographic literature and the Icelandic context, Durrenberger and I tried to separate the rhetorical element of production discourse and the realities of differential success, including the role of the skipper effect (Pálsson and Durrenberger 1982; 1983). Using methods similar to those used in Acheson's pioneering study (1977) of lobster fishing in Maine - a series of path models - we tried to assess statistically the skipper's contribution in the case of the winter fleet of Sandgerði.10 The 'residual' in our analyses, the percent of variance unaccounted for by the variables of boat size and number of trips, represented the effects of all other factors that could possibly influence catch - such as the skipper and crew, the mechanical condition of the boats, the weather, and economic and political factors. Given the small residual in our calculations, we concluded that the skipper effect was far less than generally assumed. If differential success was principally a 'material' matter, individual differences in fishing skills ('fishiness', fiskni) were overstated in folk accounts.

Several studies have addressed similar issues, attempting to identify the nature and causes of differential success by statistical means (see, for instance, Thorlindsson 1988; Gatewood and Mace 1990; Bjarnason and Thorlindsson 1993). One of the problems, however, in most, if not all, of the statistical studies involved is that they exhibit certain methodological flaws and weaknesses. As Gatewood and Mace point out (1990: 346), the findings of many of these studies typically remain ambiguous because the key 'variables' (skipper's skills, in particular) are relegated to the status of residual effects - in other words, they are not measured directly. Another problem is that some of the variables - skipper, boat size, and fishing effort - tend to be confounded; a good skipper tends to have a large boat and fish frequently and it is difficult, therefore, to separate the effects of person, effort, and technology (see Gatewood 1984).

The central issue, however, is not so much a matter of statistics and methodological refinement; rather, it has to do with the conceptual issues which remain. Take, for one thing, the 'confounded variables' of person and technology. The problematic nature of technology and its relation to the person is well captured in the Icelandic distinction sometimes made between those who fish 'by skill' (af lagni; literally, with dexterity) and those who fish 'by force' (af krafti; literally, with power). In the former case, typically involving an experienced skipper, dexterity and alertness to the tasks at hand are of primary importance; in the latter case, usually that of a novice, it is technology that counts. The skilful skipper attends to his fishing technology as if it were an extension of his person; the novice, in contrast, focuses on the gear itself - fetishising his gadgets, making more trips, and using more fuel, often destroying more gear in the process. Where are we then, to draw the boundaries between fishermen, their technology, and the environment? Polanyi observes (1958: 59) that as we become skilful practitioners we assimilate technology as a part of our own body. In the process of learning, he suggests, 'we shift outwards the points at which we make contact with the things that we observe as objects outside ourselves'; as tools become 'part of ourselves, the operating persons', we 'pour ourselves into them and assimilate them as parts of our own existence. We accept them existentially by dwelling in them' (1958: 59; cf. Ingold 1993b).

Significantly, in Iceland experienced skippers often speak of knowing the details and the patterns of the 'landscape' of the sea bottom 'as well as their fingers'. This indicates that for the skilled skipper fishing technology - the boat, electronic equipment, and fishing gear - is not to be regarded as an 'external' mediator between his person and the environment but rather as a bodily extension in quite a literal sense. Thanks to such technological extensions the experienced skipper is able to 'see' the fish, an otherwise invisible prey, and the landscape of the sea bed - much like, given the necessary training, a blind person is able to walk with the aid of a stick. Thus, as Bjarnason and Thorlindsson point out (1993: 388-9), 'skippers operating comparable boats will make different use of them in terms of effort and efficiency'. Such bodily extensions have their ethnographic parallels; skilled fieldworkers pour themselves into their pen or notebook (or computer), not so much focusing on them as dwelling in them. In the situated practice of fieldwork, our equipment, too, has a tendency to 'disappear' (see Suchman 1987: 53).

More importantly, given the present perspective, the whole framing of the issue of differential fishing skills rests on a reductionist notion of skill and the autonomous human agent, much like sociolinguistic notions of speaking and communicative competence (Pálsson 1991: 17). Assuming that enskilment is mainly the result of practical engagement with the environment, the notion of the autonomous skipper is a misleading one; whatever different contributors to the discussion of the skipper effect may have said about the size of the statistical residual, and about the amount of variance in catches explained by the skipper's skill, there has been a tendency to think of the skipper as a separate 'variable' more or less independent of social context, and to remove his actions from the sphere of social relations. Thorlindsson, who argues for the strength of the skipper effect by comparing the hierarchies of the Icelandic herring fishery and international chess, is no doubt right in suggesting that in both cases 'individual performance is [. . .] based on skill' (1988: 206), but the analogy between a highly individualistic game and fishing - equivalent to Saussure's famous analogy between chess and speaking (see Harris 1988) - tends to draw upon a culturally specific view of the craftsman and producer. When applied to fishing, the analogy of chess suggests a dubious model of the skipper, as a person operating outside society, removing his actions from his relations with crew, other skippers and their crew, and the wider community to which he belongs.11 If those who emphasise the strength of the skipper effect tend to lose sight of the social nature of the tasks involved, then statistical analyses (including my own) that purport to show the weakness of the skipper effect, relative to folk accounts, likewise fail adequately to render the collective project of fishing.

Again, indigenous discourse provides a lead. While Icelandic fishermen and boat-operators frequently underline the role of the skipper, they also emphasise the importance of 'having a good crew' (að hafa góðan mannskap). One trawler skipper I talked to used the analogy of music, referring to the crew as an instrument and the skipper as a piano player:


A good player is useless if the piano is deficient or out of tune. Likewise, a first class piano doesn't make good music if the player is no good.


Skippers' frequent statements about the significance of 'having a good crew' have something to do with modesty; a good skipper is unlikely to elaborate on his own performance or capabilities, since boasting would not be tolerated. Nevertheless, there is every reason to take seriously skippers' claims about the importance of the crew; such statements are not merely rhetoric. To adequately render the social nature of production we may have to extend the notion of the 'skipper effect' and refer as well to a 'crew effect' (Rob van Ginkel 1992, personal communication) - if not to a 'fleet effect' (White 1989). If, for the skipper, the crew is an 'instrument' analogous to his boat and fishing gear, it logically follows, given what I said above, that it, too, forms a part of the operating person of the skipper. To return to the question I posed above, concerning the relative contributions of the individual and the larger social context, fishing success is both a matter of personalities and collectivities. And this is precisely to decenter the notion of skipperhood; in order to have success in catching fish, the skipper must dwell in his crew and the larger community of practice to which they all belong.

To continue the ethnographic analogy, to find their sea legs - to learn to 'make good music', to use the skipper's words - ethnographers must pour themselves into the community they work in. While for anthropologists such a perspective of dwelling and empathy strikes a familiar note, it also runs against much accepted wisdom, including the current image of the lonely fieldworker and the alienating discourse on violence and boundaries associated with the notion of anthropology as 'cultural translation' (see Pálsson 1993: 16).


The flow and momentum of fishing

While fishing the crew performs a highly complex operation, involving a series of specialised and interrelated tasks each of which must be performed quickly and in accordance with a given schedule. Fishing with gill nets is a good example. A net boat has several units of gill nets (trossur) and on each trip the skipper will normally visit several locations, drawing the nets set during the previous trip (usually a day or two earlier) and re-setting them. Deciding on the exact location is the skipper's responsibility; the crew are not directly involved although they take an active interest (given the application of a 'share system') and, occasionally, especially when fishing is not going well, they may offer their own views. In contrast, setting the nets and drawing them are intensely co-operative tasks. Once the boat is in the right spot, the engine is set to full speed while the crew 'cast' the nets overboard in a swift operation. The key issue here is to make sure that the nets do not become entangled, so that the operation can be completed smoothly and without accident.

Drawing the nets is a slower but far more complex process. The skipper oversees the whole operation from his base on the 'bridge'. Each member of the crew occupies a specific position on the deck, the most critical one being that of the operator of the winch which draws the net. Not only is this a highly dangerous task, it sets the speed of the operation and ensures the flow of the job. The position of the winch-operator tends to be permanent and specialised; other deck-hands, in contrast, rotate relatively freely from one position to another. Clearing the net and removing the fish also demands quick hands, careful attention, and good co-ordination. Each person has constantly to adjust his or her actions to those of the other crew as well as to the speed of the net passing through their hands. If one person fails to cope with the demands of the task, the whole operation comes to a halt. Keeping stops to a minimum not only deminishes the likelihood of accidents as well as maximising free time ashore between trips, but also increases catches relative to effort (that is, per trip). Asked about the importance of co-operation among the crew, a manager of a large fishing firm (and a former successful skipper) explained:


It almost makes all the difference to have a good crew. After all, a skipper cannot achieve considerable success unless he has a good crew. Two of our boats are almost identical, but in one case there is a constant turnover of crew - as three or four deck-hands leave the boat after each fishing trip - while in the other the crew remains essentially the same. The latter boat, with its stable and experienced crew, enjoys much greater fishing success.


Since an experienced crew is the key to efficiency and safety on deck, as well as to the economic use of gear and technology, boat owners try to avoid a high turnover in personnel. Skippers, similarly, emphasise that successful fishing is only possible as long as certain interactional demands are met - a low turnover of crew and good relationships among the persons involved. Thus skippers at the top of the local hierarchy tend to have stable crews, while those at the bottom must accept inexperienced novices much of the time. Generally, a high turnover of crew is regarded as a bad sign with severe implications; it suggests low fishing effort (relatively few trips), limited catches in any one trip, low salaries, low morale, lack of prestige, and a high accident rate. Skippers point out that while every skilful deck-hand was once a novice, and while they may have to put up with one novice at any time, a 'mixed crew' with a relatively high number of inexperienced deck-hands is both a safety hazard and a social and economic handicap.

Given the importance of skilled deck-hands and a low turnover of personnel, crew membership, the result of complex negotiations, needs careful consideration. Whether or not he owns his boat or works for someone else (the larger vessels are usually company-owned while the smaller ones tend to be the property of skippers and their families), the skipper is responsible for hiring personnel - one or two mates, cook, one or two mechanics and several deck-hands, depending on the size of the boat and the kind of fishing gear used. Potential crew members are likely, of course, to keep in mind the skipper's record in fishing. It is not, however, the only relevant consideration. Some skippers are known to be arrogant, strong leaders while others have the reputation for being democratic 'nice chaps'. As one deck-hand put it:


Skippers very much differ from one another. A skipper may be good at catching fish (aflamaður), but he can be so boring that one is not willing to put up with him and work for him.


'Cultural' factors, therefore, may outweigh economic ones. Skippers, too, may be motivated by non-economic considerations. Some skippers believe that women pose particular difficulties at sea irrespective of their competence in the narrow context of fishing. Also, friendship and other personal connections may be more important to the skipper than the reported 'market' value of a deck-hand. Often, however, it is difficult to separate the economic and the cultural; in the long run, cultural considerations may become economic, as self-fulfilling prophecies.

A good skipper needs a disciplined crew, but he also needs to be a friend of those who work with him. During fishing trips, the relationship between skipper and crew tends to be authoritarian and rigid, but when they reach harbour they often 'loosen up', shifting to a more informal gear. Sometimes skippers socialise with their crew between fishing trips, strengthening social bonds and enhancing the solidarity of the crew, even though such occasions lack many of the ritual aspects Johnson describes (1979: 248) for the 'group-binding' Portuguese caldeirada (a joint meal at a local tavern). While the hierarchy of skippers is subject to some stability, the skipper's trust and reputation is not given once and for all from the moment he begins to fish, as Barth's account would lead one to believe, emphasising the 'self-confirming pattern' (Barth 1966: 10) whereby 'good' skippers get stable crews, fish independently, and land high catches (cf. Heath 1976; Durrenberger and Pálsson 1983; White 1989); rather, trust and reputation are negotiated and maintained in the context of social discourse and changing social and economic realities, in the flux and movement of boats, capital, skippers, and crews.

The fishing crew, of course, is part of a larger context. Fishermen often speak of the personnel (mannskapur) of a boat in an extended sense - including several people ashore, those who ensure efficient repairs of equipment between fishing trips and those who bait lines and take care of nets, repairing old ones or supplying new ones. Indeed, folk accounts of fishing success often emphasise the importance of good fishing gear and the diligence of the people ashore responsible for its maintenance. 'Having a good crew', therefore, means not only being able to rely on a good fishing crew, but also being provided with good 'services' (þjónusta) on land. Other ties connect the fishing crew to a series of social networks ashore, family and relatives, neighbours, and friends. Such networks are not limited to the landing port in question as they extend over a wider area. Skippers and crew get together in their respective regional organisations and unions, to negotiate shares and salaries and, generally, to defend their economic interests. Such relations significantly affect the process of recruitment, the mutual knowledge of fishermen, working conditions on board the boat, and, generally, the momentum of fishing.

The fleet is ever-present as well. While one may speak of the vessels temporarily associated with a particular landing port and nearby fishing grounds as a 'local' fleet - the 'fleet of Sandgerði', for instance - such a fleet knows no clear boundaries (cf. Acheson 1988). The fleet, in fact, is a changing constellation of boats that are registered in different towns and municipalities, and many of the skippers, crew, and boat-owners involved are permanent residents of other localities. Moreover, the fleets of different ports are hard to separate; during fishing they merge on the boundless sea. Nevertheless, the communion at sea is a very important one. Inevitably the skipper's decisions while fishing are constrained by the decisions of other skippers and by the movements of the fleet (see Durrenberger and Pálsson 1986). While deciding where to fish is largely guided by the readings of electronic equipment and by the skipper's experience of earlier fishing seasons, of no less importance is knowing what other skippers are doing, where they are likely to be, and how much they will catch. There are obvious benefits in co-operating with other skippers and crews on a daily basis while at sea, especially after a long break in fishing; sharing information on the state of the major areas saves time and fuel and each skipper gains information about fish migrations that he could not acquire on his own. As Wilson points out (1990: 14), while fishermen seek to reduce the search problem by looking for recurrent patterns in the migration and location of fish, 'the number of observations necessary to establish regularity is far too large for any single individual to acquire'. White proposes the notion of the 'fleet effect' to address this issue, suggesing that information-sharing and variable discovery techniques within a fleet 'result in larger catches than if all boats fished alone' (1989: 25). However, because skippers compete among themselves for locations, fish, crew, and prestige, they often carefully guard valuable information available to them.

One way to solve the tricky problem of co-operating under conditions of competition is to participate in an informal club of skippers, a network with relatively stable membership, thereby exchanging information on a regular and reciprocal basis: 'This method of trading information tends to maintain strong incentives for the acquisition of new knowledge and at the same time tends to coordinate and disperse information in a way that provides few benefits for free riders' (Wilson 1990: 15). Icelandic skippers refer to the participation in clubs of this kind as 'belonging to a code' (vera í kóda); club members agree on a secret code or language which enables them to freely discuss important issues (especially catches and locations) on the inter-boat radio (a device fishermen call 'the spy') without providing information to 'outsiders'. In the absence of direct, visual clues, skippers try to learn as much as possible about their colleagues (whether club members or not) through the use of the inter-boat radio. The lengthy and repetitive exchanges between skippers, and the gossip, teasing and joking they contain often seem to be simply an end in themselves, but one should not underestimate their social significance. Such exchanges not only maintain communion in the fleet (in Sandgerði, among the crews of 46 boats of varying sizes ranging from, roughly, 10 to 200 tons), they also provide subtle information on fishing. What skippers say is interpreted on the basis of an extensive prior knowledge of catch records, fishing locations and fellow fishermen. Recently, with the introduction of mobile telephones, which allow for increasing privacy in communication between club members, the role of the inter-boat radio has been much reduced.

Due to the physical seclusion of the crew while at sea, the competitive nature of fishing, and the changing and somewhat arbitrary definition of 'local' boats, it is easy to lose sight of the importance of the fleet. During my early fieldwork I was suddenly reminded of the sociality of the fleet by a search of an unusual nature - a tragic event in which a small boat with two men on board disappeared in the middle of a fishing trip, apparently due to rough seas. Quite abruptly, 'the fleet' became visible, operating as a single unit, scanning the seascape for several hours - searching, without success, for the lost boat and crew.

The reference to sociality is not to suggest that crews and fleets are best described as collective, rule-governed entities - as sets of 'markedly stereotyped, detailed roles' (Barth 1966: 10) generated on the basis of formal statuses and transactional constraints. As I remarked earlier, with reference to Rosaldo's example of the game of basketball, rules are not the essence of games from the point of view of those who play them. Nor am I arguing that crews and fleets are supra-individual byproducts of interaction and co-ordination, reified entities in the Durkheimian sense. If we were to follow the Durkheimian lead, the agency of the skipper (and any other member of crew, of course) would dissolve in a hierarchically more inclusive agency of the collectivity of crew and fleet. Given an interpretation that abandons any radical distinction between the individual and the collective, assuming that each person is an 'ensemble' of social relations, as Marx once put it, the skipper is endowed with both agency and sociality. As developmental psychologists Fischer, Rotenberg, Bullock, and Raya point out in their discussion of learning, 'the skill concept includes the person as well as the context. It is as much a mistake to leave out the person as to leave out the context' (1993: 97). A skipper's agency is inevitably constituted within the nexus of social relations, and yet his personal authority is essential for the collective project of fishing.12

The collaborative mode of practice and enskilment emphasised in the present account is increasingly the subject of anthropological discussions, usually in connection with the notion of 'apprenticeship' (see, for example, Goody 1978; Coy 1989). Many accounts of ethnographic practice, however, are excessively self-indulgent and individualistic, in the writing-culture-genre. Assuming the perspective of practice theory, and rejecting any radical distinction between experts and laypersons, it seems essential to extend the notion of apprenticeship to the work of anthropologists (Goody 1989) - and this means to move from autobiographies and narcissistic accounts of autonomous selves to social histories of fieldwork as a co-operative enterprise sustained by relationships with other people - spouses, friends, informants, and colleagues. Fieldwork, as Gudeman and Rivera emphasise (1993), is teamwork with a flow and momentum of its own - the anthropologist produces his or her ethnography with a responding people; the fieldworker is an apprentice in the collective enterprise of making an ethnography.


Apprenticeship and attentiveness

Becoming a skilful skipper - a competent member of the collectivity of a fleet - demands several years of training, both formal and informal. A prospective skipper, usually a fisherman's son, begins his career as a deck-hand, often in school holidays during early adolescence. To get a skipper's licence, however, a fisherman has to receive formal training in a specialised institution, the Marine Academy. The formal title of the 'skipper' (skipstjóri) assumes particular legally-enforced rights and obligations, emphasised and assimilated in the classroom, through two years of formal instruction. Much of the training focuses on navigation, safety measures, and the use of electronic equipment. For skippers, however, enskilment in fishing is not a matter of formal schooling and the internalisation of a stock of knowledge; rather, it is achieved through active engagement with the environment, in the broadest sense of the term. As one skipper explained,


the Marine Academy teaches interesting subjects, including meteorology. On the other hand, you learn even more by simply taking part, by living the life at sea - especially if you are young and enthusiastic.


'Real' schooling, then, is supposed to take place in actual fishing. The emphasis on 'outdoor' learning is emhasised in frequent derogatory remarks about the 'academic' learning of people who have never 'had a pee in a salty sea' (migið í saltan sjó). Questioned about the role of formal schooling, skippers often say that what takes place in the classroom is more or less futile as far as fishing skills and differential success are concerned, although they readily admit that schooling has some good points, preventing accidents and promoting proper responses in critical circumstances involving the safety of boat and crew. Even a novice fisherman, they say, with minimal experience of fishing, is likely to know more about the practicalities of fishing than the teachers of the Marine Academy. The most important items of equipment, including fish finders, continue to be modified, new models constantly enter the market, and, in any case, each tool or machine has its 'nuances' that need to be studied and established in context, in the course of fishing. No formal training can cope with this flexibility and variability in the real world. Therefore, there is little, if any, connection between school performance and fishing success.

Advice by former skippers may be useful schooling, but here, again, what counts is what the skipper does with his information on the spot, during actual fishing. On one occasion I witnessed a discussion between a novice skipper and a retired one about a 'rocky area', a particularly difficult place to fish. The experienced skipper advised the young one on how to manoeuvre the boat and the fishing gear, elaborating on the complex details of currents, fish migrations, and sea-bed features. The novice listened carefully and then commented, somewhat perplexed, 'I guess I'll figure it out when I get there'. Some years earlier the young skipper had been a deck-hand on the old man's boat. That job had different demands - relating more to what happened aboard the vessel than in the sea - and, therefore, he said, he had rarely 'attended to' (fylgdist með) the task of locating and catching fish.

Skipper education recognises the importance of situated learning. Earlier participation in fishing, as a deck-hand (háseti), is a condition for formal training, built into the teaching programme; this is to ensure minimum knowledge about the practice of fishing. Once the student in the Marine Academy has finished his formal studies and received his certificate, he must work temporarily as an apprentice - a mate (stýrimaður) - guided by a practising skipper, if he is to receive the full licence of skipperhood.13 The attitude to the mate varies from one skipper to another; as one skipper remarked, 'some skippers regard themselves as teachers trying to advice those who work with them, but others don't'. While skippers differ from one another and there is no formal economic recognition of the their role in this respect, in terms of a teaching-salary, according to many skippers the period of apprenticeship is a critical one. Reflecting on his mentor, with whom he had spent several years at sea, one skipper explained: 'I acquired my knowledge by working with this skipper, learning his way of fishing. I grew up with this man'. It is precisely here, in the role of an apprentice at sea, that the mate learns to attend to the environment as a skipper. Working as a mate under the guidance of an experienced skipper gives the novice the opportunity to develop attentiveness and self-confidence, and to establish skills at fishing and directing boat and crew. The role of the mate, in fact, institutionalises what Lave and Wenger (1991) term 'legitimate peripheral participation', a form of apprenticeship that allows for protection, experimentation, and varying degrees of skill and responsibility. This is not a one-way transfer of knowledge as the skipper frequently learns from the co-operation of his mate; mate and skipper - in fact, the whole crew - educate each other. In the beginning, the mate is just like an ordinary deck-hand; in the end he is knowledgeable enough to have a boat of his own. At first he is of little help to his tutor, later on he can be trusted with just about anything; occasionally, the skipper may even take a break and stay ashore, leaving the boat and the crew to his mate.

Often the advice of the skipper is in the form of verbal directions. He will draw the novice's attention to various aspects of skipperhood - how to manoeuvre the boat, how to use electronic equipment, how to follow fish migrations, and so on. A skipper is unlikely to share his most personal tricks, as the mate may later on become one of his competitors, in charge of another local boat. Every skipper has a personal 'diary' with details about times, species, catches, and fishing locations, and mates are rarely, if ever, given direct access to this information (some of it is stored in the memory of a computer located in the wheel-house and, significantly, passwords are sometimes used for protection). On the other hand, a keen novice will gradually learn to imitate the actions of his skipper, observing his decisions and carrying out his commands. One skipper pointed out that he 'simply' learned most of what he knew 'by seeing how others handled their tasks'.

From the point of view of the mate, the choice of a tutor is very important. In the words of one skipper, 'to get an experienced and clever man is the most important thing that can happen to a beginner, the greatest luck of all'. Some skippers envy colleagues who have had the opportunity to work early in their career with renowned skippers, acknowledged mentors and 'men of catch' (aflamenn). Sometimes, they say, the period of apprenticeship is terminated too soon. Thus, one skipper who enjoyed modest success claimed that he 'did not spend enough time as a mate' with an experienced skipper, and that he regretted having left too soon to become a full-time skipper. This was the main reason, he continued, why he could not compete with the top skippers. There may be several reasons, personal and economic, for quitting apprenticeship too early. For one thing, a retiring skipper-owner, or someone with a temporary health problem, may encourage his son to take over at an early stage, before the latter really matures as a skipper. This seems to have been the case with a young and unsuccessful skipper in one of the communities in which I worked. He was repeatedly subject to critical discussions and some even laughed at his flamboyance and inexperience; one fisherman commented, when hearing the skipper's voice on the inter-boat radio, that the skipper 'could not even handle the channels on the radio'. Another fisherman made sarcastic comments about the 'unlucky' skipper who tended to go fishing in bad weather and vice versa, staying home when the weather was good. The implication was that the skipper was inattentive. Other skippers, it seems, were reluctant to respond to the novice on the radio, despite his insistence to communicate. Skippers who do not take the period of apprenticeship seriously, or who cut it short, invariably fish 'by force' and not 'by skill'; they are referred to as göslarar, inattentive and clumsy skippers. To become a respected skipper, a fisherman not only has to become a mate and later on a skipper, he also has to 'prove himself' (sanna sig) in his capacity as skipper over several seasons, 'growing up' or 'developing himself' (verða að manni).

The skipper's knowledge is a complex one; a skipper must choose times and places to fish on the basis of a series of detailed environmental information. It is not surprising, therefore, that fishermen often refer to the importance of 'attentiveness' (eftirtekt, athygli) and 'perceptiveness' (glöggskyggni); the ability to recognise and apply an array of minute but relevant details. Attentiveness is a complex ability and includes, for example, being able to 'read' the sky and predict the weather, to participate in discussions within the local fleet, to understand the 'sparks' of electronic instruments, and to be able to co-ordinate crew activities. In some fisheries, for instance in Brazil (see Kottak 1992), technological and economic changes seem to have resulted in rapid de-skilling.14 There is little reason to believe that this has been the case in Iceland. While old and retired skippers sometimes point out that fishing has been radically transformed by electronic technology (including the computer), emphasising that 'natural signs' are increasingly redundant, attentiveness continues to be one of the central assets of the good skipper and, just as before, it demands lengthy training. The skipper's universe is very different from that of his colleagues of earlier decades, but what shows on the screens of the radar, the computer, and the fish-finder is just as much a 'natural sign', directly sensed, as birds in the air or natural landmarks.15

As in the Alaskan situation described by Gatewood (1983), the most important test of skippers' skills, and the most relevant criterion for their prestige, is fishing success. Good grades and formal schooling - and sizeable fishing quotas - are useless if one fails to catch fish. Unlike their Alaskan colleagues, on the other hand, Icelandic skippers are not expected to follow 'rational' procedures, nor do they say they necessarily do. If skippers account for their procedures at all when making decisions about the locations of prey, they are likely to say that they follow what Gatewood calls 'reasonable' procedures. Many skippers have described how a dilemma regarding fishing locations was solved by a strange message or intuition, some kind of 'whisper'. When explaining exceptional catches during single fishing trips, they sometimes claim they simply know intuitively (finna á sér) what to do. Often they also refer to 'messages' or hunches received on the spot and the state of 'fishing mood' (fiskistuð), a peculiar psychological experience - 'like being possessed' - beyond skippers' understanding and control. As one skipper pointed out,


sometimes one is so thoroughly confused when leaving harbour that one hasn't got a clue as to where to go. Then one may suddenly get a hunch.


Skippers sometimes speak of cerebral processes in this context, of 'knowledge in the head' and 'being able to think like fish', but they also refer to wisdom 'carried in the blood' and 'unexplainable' bodily judgement.

Skippers are likely to have a plan when they embark on a fishing trip, but in the process of fishing they may be forced to revise it, or even to abandon it:


Sometimes one tries to follow some plan, based on one's experience from the day before or the recent past. And then a new day arrives, with new kinds of circumstances, and it simply becomes impossible to follow a plan.


While fishing, the skipper's tactics are continuously adapted to new circumstances - the teamwork on board, the decisions of other skippers, the amount and kinds of fish caught, weather forecasts, and ocean movements.

Although skippers frequently discuss the relevance of attentiveness, they rarely mention how they actually make decisions. One reason is that they are guided more by practical results than by an interest in theoretical advancement. Often they 'simply' notice that a particular strategy seems to work, without worrying about why that is the case. Skippers' reluctance to discuss their own fishing tactics is not only the result of competition and secrecy, and the value they generally place on independence and modesty; they have difficulties in verbalising their complex experience and intuition. What fishermen label as hunches and fishing mood is particularly difficult to verbalise; some important decisions are 'out of the blue'. What matters most of all, one skipper explained, is 'how one's body will cope'. Decision-making, then, is based less on detached calculation or 'mental' reflection than on practical involvement. Similarly, the process of enskilment is not just a cognitive one; rather it involves the whole person interacting with the social and natural environment.

Once again, the analogy with ethnographic practice is revealing. Gatewood (1985) has told the story of how he became a competent deck-hand in salmon seining in Alaska. After only a few days he had to abandon his 'arrogant' hopes of being a participant-observer; trying to cope with the demands of fishing did not leave much time and energy for anything else - 'I had all I could handle', he says (Gatewood 1985: 205), 'trying to participate' - and besides, the cognitive demands of 'observation' were quite irrelevant, if not detrimental, to the learning process. Learning was not a matter of verbal encoding and enculturation but of adapting to the 'practical constraints' of the job (Gatewood 1985: 216):


The development of a seiner's cognitive organization is directed not by some transcendental need to share meanings but by the practical constraints of coordinating his actions with those of his fellows. His actions, and theirs, speak louder than words.


Edelman's ethnographic study (1993) of shunting in a Swedish railway yard provides similar observations. At the beginning of her fieldwork Edelman was preoccupied with the 'middle-class' notion of personal autonomy, of distinguishing herself from others, a notion which proved to be an obstacle to the progress of her fieldwork. Later on, however, she was 'beset . . . by the reciprocity of the demands and obligations that spring from the daily toil, the constant efforts to co-operate and create a "flow" in the work, the logic of the work-process, and the meanings that were attached to individual actions' (1993: 160-1).16

As a novice fieldworker and fisherman I, too, could not help noticing the importance of co-operation, of co-ordinating one's bodily movements and the actions of others to ensure the continuity of the enterprise. Such mutual attentiveness, the result of collective enskilment, is essential for efficient team-work, the synchronisation of the tasks involved. Occasionally, the Icelandic skipper will shout commands at his crew, over the noise of the engines, during the process of drawing a net: 'loosen the net', or 'stop the winch'. The better the crew know each other, however, the more likely it is that they will be able to predict one another's actions and respond appropriately in emergencies, and the less likely is the skipper to intervene. Silence is a sign of smooth co-operation and, likewise, direct commands signify the opposite - low morale. Most of the time, words are beside the point simply because actions speak faster than words.



Icelanders, as we have seen, sometimes apply the metaphor of the journey, the fishing trip, to the issue of personal enskilment. This is to suggest that learning is not a purely cognitive or cerebral process, a mental reflection on differences in time and space, but is rather grounded in the contexts of practice, involvement and personal engagement. Enskilment is not only likened to the physical experience of seasickness, it is indeed a bodily exercise. To become skilled at something like catching fish is to progress from nausea to well-being, to feel at home in both one's body and the company of others. As Bourdieu reminds us, socialisation produces very real, physical effects, durable dispositions inscribed in the habitus. The novice imitates the actions of others, not simply their models and discourses; practical schemes 'pass directly from practice to practice without moving through discourse and consciousness' (Bourdieu 1990: 74).

Another argument from Icelandic discourse, concerning the situatedness of decisions and the importance of team work, poses a fundamental challenge to the terms of academic debate on differential success, in particular the 'skipper effect'. Fishing, after all, is a social enterprise. While the skipper is unlikely to acknowledge, let alone ask for, the advice of his crew and other skippers, and although it may be tempting to think of his decisions as independent ones and more or less his own, each of them is constrained by the context in which it is made, much like a discourse is inevitably constrained by context, including what has been said before. In fact, we may well think of a fishing season as a 'long conversation' involving a whole community of actors - both at sea and ashore - in which the agency of each is continually constituted in relation to the others. Folk accounts of the context of decision-making and the importance of the crew and the local fleet indicate an indigenous practice-theory of fishing, emphasising that what skippers do and what happens at sea is not the result of an internalised system of formal decision rules, as normative learning theory assumes, but of hunches and tacit knowledge, of insights developed on the spot but firmly grounded in the flux and collectivity of fishing. For those familiar with the mood and momentum of a game like basketball - the rhythm of a team and the flux of physical movement and bodily contact - the practice perspective of fishing immediately has parallels. As Larry Bird, a superb basketball player, has reasoned, '[a lot of the] things I do on the court are just reactions to situations. [. . .] I don't think about some of the things I'm trying to do. [. . .] A lot of times, I've passed the basketball and not realized I've passed it until a moment or so later' (cited in Dreyfus 1991: 93). Skills - in fishing or doing fieldwork (or anything else, for that matter) - are indeed individual in the sense that they are properties of the body, dispositions of the habitus. However, to isolate their acquisition and application from everything outside the boundaries of their soma is to subscribe to a normative theory of learning and a natural conception of the individual. An alternative approach recognises the sociality of the individual being and the situated nature of human activities. If, as Bakhtin has argued (1986: 293-4), every word in conversation is half someone else's, every fish that gets caught is partly that of others.

It is one thing to participate in face-to-face interaction and to engage in fieldwork and quite another, perhaps, to 'write it up'. What would the perspective of practice theory adopted here imply for the latter project - the construction of comparative knowledge or ethnographic writing outside the field? Bloch emphasises (1991: 193) that anthropologists must be aware that by writing up the knowledge of the people they study they are not merely reproducing it but rather 'transmuting it into an entirely different logical form'. How can we, then, textualise the knowledge we acquire in the field without - in the process - fundamentally distorting both its character and the manner in which it was acquired? The important issue is not, however, whether or not our models agree with the models of the actors, but the way in which we continue the ethnographic conversation, the character of the social relations of ethnographic production. It may be easy to labour under the Orientalist illusion - assuming, as Knowlton puts it, that we are able 'to keep the "native" safely in our notes or at the end of a journey, rather than among us' - since, thereby, 'we avoid having our caricatures challenged and our categories confused in the messiness of solidary sociality' (Knowlton 1992: 78). By terminating the ethnographic conversation once we are 'out of the field', we preserve our essentialist enterprise, localising our anthropological voices rather than anthropologising local voices. Arguably, however, we never actually leave the field as long as we take part in the ethnographic enterprise; in one way or another we keep on conversing with the people we study, either directly (especially nowadays) as anthropologists 'at home' or as visitors returning to a host, or, indirectly, through inner talk.

If the act of writing up is inevitably situated in a boundless community of practice - a community that admits no radical distinction between modellers and participants, scholars and natives, scientists and informants - then indeed ethnography is best regarded as a 'perpetual discussion' (Gudeman and Rivera 1990: 4). Just as it is important to go beyond the 'claustrophobic' concepts of cognition characteristic of normative learning theory (Lave 1988; Reed 1993: 63), it is important to go beyond the monologic notion of ethnographic production. In such a view, ethnography becomes, to paraphrase Benhabib (1992: 9), a 'moral conversation exercising enlarged thinking'. As ethnographers, we may be nauseated at times, given the shifting grounds of post-modernity and the contemporary critiques of monopolistic narratives and totalising accounts. However, only by engaging ourselves in a continuous dialogue - with both our 'informants' and fellow ethnographers - can we realistically expect to find our ethnographic sea legs.



The ethnographic material presented here derives from fieldwork in Icelandic fishing communities in 1979, 1981, and 1993. In revising this article I have benefited extensively from the comments and suggestions of Hastings Donnan (the editor of Man), Tim Ingold, and three anonymous reviewers. Several other scholars have also been of help - including Myrdene Anderson, E. Paul Durrenberger, Stephen Gudeman, Guðný Guðbjörnsdóttir, Stephen A. Marglin, Bonnie McCay, Halldór Stefánsson, Rob van Ginkel, and Torben Vestergaard. Agnar Helgason provided valuable assistance with interviews. Torfi H. Tulinius translated the French abstract. An earlier version of the article was presented at a work-shop on the 'Anthropology of Iceland' at the University of Iowa (28-30 May 1993) and a symposium on 'Anthropology and Nature' at the University of the Faeroe Isles (27-31 October 1993). The research on which the article is based has been supported by several funds and institutions, including the Nordic Environmental Research Programme 1993-1997, the Nordic Committee for Social Science Research (NOS-S), the University of Iceland, the Icelandic Science Foundation, and the British Council.


1. Given the perspective of programmatic knowledge embodied in cultural models, there is a close convergence between cognitive anthropology, on the one hand, and, on the other hand, artificial intelligence and expert systems (Suchman 1987: 7-10); just as programmers write 'cultures' for robots, some cognitive anthropologists endeavour to write cultural grammars of everything a competent 'native' has to know.


2. Some scholars, including Lave (1988), speak of 'functionalist' theories of learning when referring to what I call 'normative' approaches. In the present context, the former label would be somewhat misleading, since for most anthropologists the notion of the 'functionalist' school of thought has a definite and much broader meaning.


3. Marglin uses the example of billiards which many economists, including Milton Friedman, have found particularly compelling. Friedman suggests (1953) that when lining up a delicate shot, outstanding billiard players behave as if they were experts in differential equations; they need not try to solve the complicated equations of the interactions of the billiard balls, but they are necessarily governed by utility maximization, acting as mathematicians. In Marglin's terms, this is to wrongly identify the players' tacit knowledge with episteme: 'Friedman has no warrant - apart from the weight of Western intellectual tradition - to reduce the billiard player's techne to the mathematician's episteme. In the present perspective, her knowledge is of an entirely different kind, belonging to a different system, and it is no more instructive to suggest that the billiard player lines up her shots "as if" she were a mathematician than to suggest that the mathematician solves her equations "as if" she were trying to make a score at the billiard table' (Marglin 1990: 240-41).


4. Thus, in the management regime of Icelandic fishing, which is largely informed by professional marine biologists, there is very little attempt to utilise the extensive knowledge that skippers have achieved after years of practical experience of fishing; skippers frequently complain that marine biologists tend to treat them 'as idiots'. The present study is part of a larger, collaborative research project, 'Common Property and Environmental Policy in Comparative Perspective', one of the aims of which is to explore how indigenous producers acquire knowledge about the ecosystem within which they operate, how their knowledge differs from that of professional biologists, and to what extent the former could be brought more systematically into the process of resource management.


5. Malinowski, Leach suggests, 'specifically mocked at the account of Trobriand social structure that one might expect to obtain from a professional Trobriand informant, . . . though when he himself attempted to write a concise description of "The Constitution of Trobriand Society" . . . the result resembles most strikingly that given by his imaginary despised "informant"' (1970: 134-5).


6. Hastrup's argument stresses the radical distinction between natives and ethnographers, and the inevitable superiority of the 'scientific' discourse of the latter: 'At the autobiographical level ethnographers and informants are equals; but at the level of the anthropological discourse their relationship is hierarchical. . . We select the quotations and edit the statements. . .' (Hastrup 1992; 122, emphasis in the original). 'Because any scientific discourse must make claims to speak over and above the acts observed or heard', Hastrup goes on, 'there is an inherent hierarchy in the relationship between the interlocutors. To deny that is also to remain insensitive to the violence inherent in fieldwork' (1992: 122, emphasis added). Einarsson suggests (1990), on the other hand, that the 'native point of view' is misrepresented in Hastrup's structural analyses and that they over-interpret the evidence they deal with.


7. We may be able to show that publicly available information on immediately past fishing activity systematically relates to skippers' decisions with respect to the timing and location of fishing, but that does not tell us how such information enters decisions (Durrenberger and Pálsson 1986: 216)


8. It is not quite clear what Sahlins's reference to 'common sense and experience' should be taken to mean. For some fisherfolk, differences in catches are relatively straightforward, for others they are an elusive matter involving magic and sorcery. Thus, in his analysis of peasant fishing in Sweden Löfgren emphasises that 'fishing luck' is regarded as an individual resource which can be manipulated by others; for him, luck is one element of 'the "little tradition" of sorcery' (1989: 52) related to the ethos of the limited good. Many anthropologists, one may add, have been willing to entertain the idea that actual differences are undercommunicated in indigenous discourse; there are numerous examples in the ethnographic literature on hunter-gatherers - see, for instance, Lee (1988) on 'primitive communism'. The opposite case, on the other hand, of overstated differences, has received relatively little attention, at least in the literature on fishing.


9. Such a model is in stark contrast to the model of the household economy of earlier centuries, an egalitarian model that tended to minimise differences among producers (see Pálsson 1991: Ch. 4). The modern skipper is a 'strong' leader fighting for his place in the hierarchy of prestige, whereas the 'foreman' (formaður) in charge of the rowing-boat during the era of the peasant economy was a 'weak' one, or simply one among equals. While modern fishing records are public documents which compare the catches of different skippers or boats for the same season (aflaskýrslur; literally, records of catches), the records of the past were personal or private ones, comparing the number of trips of a particular boat over several seasons (róðratöl; literally, records of trips).


10. We found, much to our surprise, that our path models and regression analyses could account for all but little (about 15%) of the variance in the catch by means of two independent variables: boat size and number of fishing trips. As this did not leave much room for the skipper effect we expected to see, we repeated this exercise for several years and places. The results, however, were similar each time - with the exception of herring fishing in summer which is an entirely different matter, for ecological as well as technical reasons (Durrenberger and Pálsson 1983).


11. The analogy with chess, frequently referred to in studies of expertise (Bloch 1991: 188), need not, however, suggest an autonomous individual. Even a 'solitary' game like chess is an interactive phenomenon constituted within a social and historical context. The player, much like the skipper, is carried by the momentum and the flow of the game, mutually constituted by the participants and their psychologies, the judges, and their rules - in short, a community of practice situated in the political economy of chess.


12. Interestingly, in his 'tearing apart' of the notions of directing authority and individuality, in the chapter on co-operation in Capital, Marx (1976[1867]: 443) seems to adopt a Durkheimian understanding of the sociality of the fishing crew. When people co-operate 'in a planned way with others', he argues, 'as . . . in fishing for herrings' (1976: 445), the worker 'strips off the fetters of his individuality' (1976: 447), thereby creating 'a new productive power, which is intrinsically a collective one' (1976: 443).


13. The present discussion is obviously male-centred; although on Icelandic fishing vessels women occasionally work as cooks and deck-hands, they are rarely employed as mates or mechanics and never, as far as I know, as skippers.


14. Kottak argues that the personal skills (eyesight and landmark memory) that once were said to distinguish among fishing captains became redundant as a result of motorisation: 'So unnecessary had the captain's traditional skills become that even men who had never fished before bought motorboats and served as captains. . . . These captains were now drawing a higher income because of their property rather than their special skills' (1992: 130). Löfgren (1977) and Byron (1986) also discuss the social implications of radical technological and economic change in fishing - in Sweden and Shetland, respectively.


15. As I argued above, however, the quota system is currently transforming the ways in which Icelanders talk about personal differences in productivity and fishing success; differential access to the means of production is changing indigenous discourses on the scope of competition and the nature of production. The owners of quotas - the 'lords of the sea' (sægreifar), as they are sometimes called in heated debates on fisheries management - are taking the centre-stage, pushing the heroic skipper to the margin. Skippers, and fishermen generally, increasingly employ a feudal metaphor when discussing their own predicament, referring to themselves as the 'tenants' (leiguliðar) of quota holders.


16. Interestingly, while at first Edelman tended to assume that shunting skills were assessed only on the basis of individual achievement, later on she realised that, for the shunters, sociality and co-operation were of no less importance. Although shunting skills, she suggests, are individual in the sense that they are manifested in the body - engraved in the 'spinal marrow', as the shunters put it - in fact they cannot be separated from the flow of shunting, from participation and interaction.



Acheson, J.M. 1977. Technical skills and fishing success in the Maine lobster industry. Human Ecology 3, 183-207.

---- 1981. Anthropology of fishing. Annual Review of Anthropology 10, 275-316.

---- 1988. The lobster gangs of Maine. Hanover & London: University Press of New England.

Bakhtin, M.M. 1986. Speech genres and other late essays. Transl. V.W. McGee, eds C. Emerson & M. Holmquist. Austin: Univ. of Texas Press.

Barth, F. 1966. Models of social organization. Occasional Paper No. 23. London: Royal Anthropological Institute of Great Britain and Ireland.

Benhabib, S. 1992. Situating the self: gender, community and postmodernism in contemporary ethics. Cambridge: Polity Press.

Bjarnason, Th. & Th. Thorlindsson 1993. In defense of a folk model: the 'skipper effect' in the Icelandic cod fishery. American Anthropologist, 95, 371-94.

Bloch, M. 1991. Language, anthropology and cognitive science. Man 26, 183-98.

Bourdieu, P. 1990. The logic of practice. Trans. R. Nice. Cambridge: Polity Press.

Byron, R. 1980. Skippers and strategies: leadership and innovation in Shetland fishing crews. Human Organization 39, 227-32.

----- 1986. Sea change: a Shetland society, 1970-1979. St. John's: Institute of Social and Economic Research.

Coy, M. (ed.) 1989. Apprenticeship: from theory to method and back again. New York: SUNY.

Dreyfus, H.L. 1991. Being-in-the-world: a commentary on Heidegger's Being and time, division I. Cambridge, Mass.: The MIT Press.

Durrenberger, E. P. 1993. The skipper effect and folk models of the skipper effect among Mississippi shrimpers. Human Organization 52, 194-202.

Durrenberger, E. P. & G. Pálsson 1983. Riddles of herring and rhetorics of success. Journal of Anthropological Research 39, 323-35.

----- 1986. Finding fish: the tactics of Icelandic fishermen. American Etnologist 13, 213-29.

----- 1987. Ownership at sea: fishing territories and access to sea resources. American Ethnologist 14, 508-22.

Edelman, B. 1993. Acting cool and being safe: the definition of skill in a Swedish railway yard. In Beyond boundaries: understanding, translation and anthropological discourse (ed.) G. Pálsson. Oxford: Berg.

Einarsson, Níels 1990. From the native's point of view: some comments on the anthropology of Iceland. Antropologiska Studier 46-47, 69-77.

Ellen, R. 1982. Environment, subsistence and system: the ecology of small-scale social formations. Cambridge: Univ. Press.

Fischer, K.W., E.J. Rotenberg, D.H. Bullock, & P. Raya 1993. The dynamics of competence: how context contributes directly to skill. In Development in context: acting and thinking in specific environments (eds) R.H. Wozniak & K.W. Fischer. Hillsdale, NJ: Lawrence Erlbaum Associates.

Friedman, M. 1953. The methodology of positive economics. In Essays in positive economics. Chicago: Univ. of Chicago Press.

Gatewood, J.B. 1983. Deciding where to fish: the skippers' dilemma in Southeast Alaskan salmon seining. Coastal Zone Management Journal 10, 347-67.

----- 1984. Is the 'skipper effect' really a false ideology? American Ethnologist 11, 378-9.

----- 1985. Actions speak louder than words. In Directions in cognitive anthropology (ed.) J.W.D. Dougherty. Urbana & Chicago: Univ. of Illinois Press.

Gatewood, J. & P.M. Mace 1990. Hard work versus patience: simulation of fishing strategies in Nova Scotian herring purse seining. Journal of Quantitative Anthropology 2, 329-52.

Gibson, J.J. 1979. The ecological approach to visual perception. Boston: Houghton-Mifflin.

Gladwin, T. 1964. Culture and logical process. In Explorations in cultural anthropology (ed.) W.H. Goodenough. New York: McGraw Hill.

Goody, E.N. 1978. Towards a theory of questions. In Questions and politeness: strategies in social interaction (ed.) E. Goody. Cambridge: Univ. Press.

----- 1989. Learning, apprenticeship and the division of labor. In Apprenticeship: from theory to method and back again (ed.) M.W. Coy. Albany: State Univ. of New York Press.

Gudeman, S. & A. Rivera 1990. Conversations in Colombia: the domestic economy in life and text. Cambridge: Univ. Press.

----- 1993. From car to house. Paper presented at the annual meeting of the American Anthropological Association, Washington, DC, 17-20 Nov.

Harris, R. 1988. Language, Saussure and Wittgenstein: how to play games with words. London: Routledge.

Hastrup, K. 1992. Writing ethnography: state of the art. In Anthropology and autobiography (eds) J. Okely & H. Callaway. London & New York: Routledge.

----- 1993. The native voice - and the anthropological vision. Social Anthropology 1, 173-86.

Heath, A.F. 1976. Decision making and transactional theory. In Transaction and meaning: directions in anthropology of exchange and symbolic behavior (ed.) B. Kapferer. Philadelphia: Inst. for the Study of Human Issues.

Holland, D. & N. Quinn 1987. Culture and cognition. In Cultural models in language and thought (eds) D. Holland & N. Quinn. Cambridge: Univ. Press.

Hutchins, E. 1983. Understanding Micronesian navigation. In Mental models (eds) D. Genter & A. Stevens. Hillsdale, NJ: Erlbaum.

Ingold, T. 1993a. Tool use, sociality and intelligence. In Tools, language and cognition in human evolution (eds) K.R. Gibson & T. Ingold. Cambridge: Univ. Press.

----- 1993b. Technology, language, intelligence: a reconsideration of basic concepts. In Tools, language and cognition in human evolution (eds) K.R. Gibson & T. Ingold. Cambridge: Univ. Press.

Joas, H. 1993. Pragmatism and social theory. Chicago & London: The Univ. of Chicago Press.

Johnson, T. 1979. Work together, eat together: conflict and conflict management in a Portuguese fishing village. In North Atlantic maritime cultures: anthropological essays on changing adaptations (ed.) R. Andersen. The Hague: Mouton Publishers.

Knowlton, D. 1992. No one can serve two masters or native anthropologist as oxymoron. International Journal of Moral and Social Studies 7, 72-88.

Kottak, C.P. 1992. Assault on Paradise: social change in a Brazilian village. Second ed. New York: McGraw Hill.

Lave, J. 1988. Cognition in practice: mind, mathematics and culture in everyday life. Cambridge: Univ. Press.

----- 1990. The culture of acquisition and the practice of understanding. In Cultural pychology: esays on comparative human development (eds) J.W. Stigler, R.A. Shweder & G. Herdt. Cambridge: Univ. Press.

Lave, J. & E. Wenger 1991. Situated learning: legitimate peripheral participation. Cambridge: Univ. Press.

Leach, E.R. 1970. The epistemological background to Malinowski's empiricism. In Man and culture: an evaluation of the work of Malinowski (ed.) R. Firth. London: Routledge & Kegan Paul.

Lee, R.B. 1988. Reflections on primitive communism. In Hunters and gatherers: history, evolution and social change (eds) T. Ingold, D. Riches, & J. Woodburn. Oxford: Berg.

Levieil, D.P. & B. Orlove 1990. Local control of aquatic resources: community and ecology in Lake Titicaca, Peru. American Anthropologist 92, 362-82.

Lock, M. 1993. Cultivating the body: anthropology and epistemologies of bodily practice and knowledge. Annual Review of Anthropology 22, 133-55.

Löfgren, O. 1977. Fangstman i industrisamhellet. Lund: Liber Laromedel.

----- 1989. The reluctant competitors: fisherman's luck in two Swedish maritime settings. Maritime Anthropological Studies 2, 34-58.

McGoodwin, J.R. 1990. Crisis in the world's fisheries: people, problems, and policies. Stanford: Stanford University Press.

Marglin, S.A. 1990. Losing touch: the cultural conditions of worker accommodation and resistance. In Dominating knowledge: development, culture and resistance (eds) F.A. Marglin & S.A. Marglin. Oxford: Clarendon Press.

Marx, K. 1976[1867]. Capital, vol. 1. B. Fowkes, transl.. Middlesex: Penguin.

Miller, G.A., E. Galanter, & K.H. Pribram 1960. Plans and the structure of behavior. New York: Holt, Rinehart & Winston.

Miller, M.L. & J. van Maanen 1979. 'Boats don't fish, people do': some ethnographic notes on the federal management of fisheries in Gloucester. Human Organization 38, 377-85.

Ochs, E. 1988. Culture and language development: language acquisition and language socialization in a Samoan village. Cambridge: Univ. Press.

Okely, J. 1992. Anthropology and autobiography: participatory experience and embodied knowledge. In Anthropology and autobiography (eds) J. Okely & H. Callaway. London: Routledge.

Ortner, S.B. 1984. Theory in anthropology since the sixties. Comparative Studies in Society and History 26, 126-66.

Pálsson, G. 1991. Coastal economies, cultural accounts: human ecology and Icelandic discourse. Manchester & New York: Manchester Univ. Press.

----- 1993. Introduction: beyond boundaries. In Beyond boundaries: understanding, translation and anthropological discourse (ed.) G. Pálsson. Oxford: Berg.

----- 1994. The textual life of savants: ethnography, Iceland, and the linguistic turn. New York: Harwood Academic Publishers. (In press).

Pálsson, G. & E. P. Durrenberger 1982. To dream of fish: the causes of Icelandic skippers' fishing success. Journal of Anthropological Research 38, 227-42.

----- 1983. Icelandic foremen and skippers: the structure and evolution of a folk model. American Ethnologist 10, 511-28.

----- 1990. Systems of production and social discourse: the skipper effect revisited. American Anthropologist 92, 130-41.

----- 1992. Icelandic dialogues: individual differences in indigenous discourse. Journal of Anthropological Research 48, 301-16.

Polanyi, M. 1958. Personal knowledge: towards a post-critical philosophy. Chicago: Univ. of Chicago Press.

Reed, E.S. 1993. The intention to use a specific affordance: a conceptual framework for psychology. In Development in context: acting and thinking in specific environments (eds) R.H. Wozniak & K.W. Fischer. Hillsdale, NJ: Lawrence Erlbaum Associates.

Resnick, L.B. 1987. Learning in school and out. Educational Researcher 16, 13-20.

Rosaldo, R. 1980. Ilongot head hunting 1883-1974: a study in society and history. Stanford: Stanford University Press.

----- 1987. Where objectivity lies: the rhetoric of anthropology. In The rhetorics of the human sciences (eds) J.S. Nelson, A. Megill, & D.N. McCloskey. Madison, Wisc.: The Univ. of Wisconsin Press.

Sahlins, M. 1972. Stone age economics. London: Tavistock.

Suchman, L. 1987. Plans and situated actions: the problem of human-machine communication. Cambridge: Univ. Press.

Thorlindsson, Th. 1988. The skipper effect in the Icelandic herring fishery. Human Organization 47, 199-212.

Veblen, T. 1964[1914]. The instinct of workmanship and the state of the industrial arts. Introd. by J. Dorfman. New York: Sentry Press.

Wengle, J.L. 1988. Ethnographers in the field: the psychology of research. Tuscaloosa & London: The Univ. of Alabama Press.

White, D.R.M. 1989. Knocking 'em dead: Alabama shrimp boats and the 'fleet effect'. In Marine resource utilization: a conference on social science issues (eds) J.S. Thomas, L. Maril, & E.P. Durrenberger. Mobile: Univ. of South Alabama College of Arts and Sciences Publication Vol. 1 & the Mississippi-Alabama Sea Grant Consortium.

Wilson, J.A. 1990. Fishing for knowledge, Land Economics 66, 12-29.



L'acquistion de savoirs-faire en mer


Je tente de démontrer, à partir de mon expérience de la pêche en Islande, que l'acquisition personnelle de savoir-faire, à la fois dans la pêche et dans le travail ethnographique, ne signifie pas l'intégration mécanique d'un corpus de connaissances mais plutôt une interaction active du sujet avec un environnement. Une telle perspective qui se nourrit aussi de théories de la pratique a des correspondances dans quelques aspects du discours islandais; lorsqu'il est question de l'acquisition de savoir-faire, les Islandais font parfois référence à des voyages nauséabonds sur une mer mauvaise, représentant l'apprentissage comme le processus de s'amariner (að sjóast). Je suggère que la perspective d'une théorie de la pratique a des implications importantes en ce qui concerne les discussions sur la production économique et la variabilité des performances, en particulier le débat sur 'l'effet du capitaine' (skipper effect). Les savoir-faire sont en effet des dispositions corporelles à distribution différentielle, mais isoler leur acquisition et leur application de tout ce qui est en dehors des frontières du corps équivaut à souscrire à une théorie réductionniste à la fois de l'apprentissage et de la sociabilité. La vue alternative d'une théorie de la pratique met l'accent sur l'importance d'être attentif à la personne entière, aux relations de maître et apprenti, et à la communauté plus large à laquelle ils appartiennent ce qui décentre l'analyse de l'acquisition des savoir-faire.


Faculty of Social Science, University of Iceland, 101 Reykjavík, Iceland