Alfred Schutz, phenomenology and research methodology for information behaviour research
Professor T.D. Wilson
University of Sheffield
Explores the phenomenological sociology of Alfred Schutz, with particular reference to his concern to understand the social distribution of knowledge in society. The methodological principles of phenomenology are discussed, with particular reference to 'ideal types'. Schutz's identification of the stratification of knowledge with three key ideal types is elaborated and the implications for information behaviour research are discussed. The relationship between methodology and method is discussed, and a novel typology of social research methods is offered.
Phenomenology, per se, is a branch of philosophy, owing its origin to the work of Husserl  and later writers [e.g., Heidegger, Sartre, Merleau-Ponty, who took the ideas into existentialism]. The aim of phenomenology, as propounded by Husserl, is to study human phenomena without considering questions of their causes, their objective reality, or even their appearances. The aim is to study how human phenomena are experienced in consciousness, in cognitive and perceptual acts, as well as how they may be valued or appreciated aesthetically. Phenomenology seeks to understand how persons construct meaning and a key concept is intersubjectivity. Our experience of the world, upon which our thoughts about the world are based, is intersubjective because we experience the world with and through others. Whatever meaning we create has its roots in human actions, and the totality of social artefacts and cultural objects is grounded in human activity.
Not surprisingly, phenomenology was attractive to psychological investigators in the early part of the 20th century and phenomenological psychology (alternatively called existential psychology or existential phenomenological psychology) exists as a thriving sub-discipline of psychology (Frankl, May, Perl), where the emphasis is on understanding a person's experience of a problematic situation. 'The people in question tell their own story, in their own terms. So "fidelity to the phenomenon as it is lived" means apprehending and understanding it in the lived context of the person living through the situation.' 
ALFRED SCHUTZ AND PHENOMENOLOGICAL SOCIOLOGY
Alfred Schutz was born in Vienna in 1899 and died in New York in 1959, and was remarkable in the fact that he pursued a career in banking  at the same time as pursuing his interests in phenomenological philosophy and the creation of a phenomenological basis for the social sciences - carried on through part-time work at the New School for Social Research in New York. It is, perhaps, this direct contact with the world of work outside the academy and with the day-to-day concerns of earning a living, along with a capability of observing the ordinary lives of people with whom he came into contact, that make some of his essays so readable and the ideas so accessible.
Although Schutz was never a student of Husserl (he studied law at the University of Vienna, on his return from military service in Italy during the First World War), he, together with a colleague, Felix Kaufman, studied Husserl's work intensively in seeking a basis for a 'sociology of understanding' derived from the work of Max Weber . This work and its continuation resulted in his first book, Der sinnhafte Aufbau der sozialen Welt  (literally, 'The meaningful construction of the social world', but published in English as 'The phenomenology of the social world' ). This work brought him to the attention of Husserl, with whom he corresponded and whom he visited until Husserl's death in 1938. In fact, he was offered the position of assistant to Husserl at Freiburg University in the early 1930s, but declined.
SCHUTZ'S METHODOLOGICAL POSITION
Schutz's methodological position is set out in three essays in Volume 1 of his Collected Papers . His starting point is that social research differs from research in the physical sciences by virtue of the fact that, in the social sciences, one is dealing with 'research objects' that are themselves interpreting the social world that we, as scientists, also wish to interpret. People are engaged in an on-going process of making sense of the world, in interaction with their fellows and we, as scientists, are seeking to make sense of their sense-making. In doing so, we must inevitably make use of the same methods of interpretation as does the person in his or her 'common-sense world'. What distinguishes the social scientific enterprise, however, is that the social scientist assumes the position of the disinterested observer. He or she is not involved in the life of those observed - their activities are not of any practical interest, but only of cognitive interest.
In Schutz's words, the ordinary person, acting in the world, is in a biographically-determined situation, doing what he or she does according to the system of relevances that enables them to select from the environment and from interactions with others, those elements that make sense for the purpose at hand. The social scientist, on the other hand, is operating on the basis of a scientifically-determined set of relevances, choosing those aspects of the situation that are appropriate for the objectives of the research. As a result, the social scientist may focus on aspects of behaviour that are taken-for-granted by the ordinary person, but that are topics of cognitive interest to the social scientist.
In behaving in this way, the researcher develops models of human action, the general postulates of which are,
- the postulate of logical consistency, whereby the objective validity of the scientist's constructs are guaranteed and are distinguished from the constructs of everyday life;
- the postulate of subjective interpretation, whereby the scientist can refer '...all kinds of human action or their result to the subjective meaning such action or result of an action had for the actor." , and
- the postulate of adequacy: that is, the constructs created by the researcher should be understandable by the individual social actor and his/her fellows. Compliance with this postulate ensures that the scientific constructs are consistent with the constructs of common-sense experience of the social world.
The models of human action are created through a process of typification, a concept found in both Weber and Husserl, and derived by Shutz from both. He first encountered the idea in Weber's concept of 'ideal types', and then in Husserl's insight that typification is a key process in our sense-making about the world. Our common-sense typifications are used continuously and may be under continuous development, while scientific typifications - social types (i.e., types of actors, types of actions, types of social personalities) in the case of Schutz, serve the same kind of purpose, but within the relevance structures and research objectives of the social scientist. They provide a means of identifying, classifying and comparing modes of social action and interaction, using defined criteria for the assignment of phenomena to type.
TYPES EXEMPLIFIED: 'THE WELL-INFORMED CITIZEN'
One of Shutz's essays  exemplifies the idea of types, and presents us with an opportunity to consider the value of the idea for information science. 'The Well-informed Citizen', is actually a contribution to the sociology of knowledge, as its sub-titled suggests - 'an essay on the social distribution of knowledge'. However, Schutz was taking issue with the Marxian sociology of knowledge as expressed by Mannheim , which represented knowledge as a product of material social conditions. Schutz, on the other hand, demonstrates that knowledge is derived from people's practical experience of the world.
He does this by constructing the ideal types of 'the man on the street', 'the citizen who aims at being well-informed', and 'the expert'. Knowledge is socially distributed according to these types and by what is taken for granted by each, and according to what system of relevances. Thus, the 'man on the street' operates according to his (or her) set of naive relevances, for which 'recipe knowledge' is adequate. I, as baker, need only a recipe for bread making in order to accomplish an eatable loaf of bread - I do not need to know the chemistry of yeast or any other scientific aspect of bread-making. Those aspects are not relevant to my immediate purpose. In another essay, 'The stranger' Schutz notes that:
'…the knowledge of the man who acts and thinks within the world of his daily life is not homogeneous; it is (1) incoherent, (2) only partially clear, and (3) not at all free from contradictions.' 
We might well conclude that this is also the case for the other two ideal types.
I, as election time draws near, will seek to make myself well-informed as to the choices before me that are relevant to me both as citizen, with civic responsibilities, and as private person, with needs that the political parties may seek to address differently. The fact that I take this decision at this time suggests that the 'zone of relevance' for my information-seeking activities as 'well-informed citizen' is not constant, but will vary according to my interests.
I, as an expert (to some degree at least) in this small sub-area of information science (or of sociology, if we wish to define it as such) seek fundamental knowledge that will lead to an understanding of behaviour. Schutz adds that the system of relevances according to which I work as 'expert' is imposed by the pre-established problems of the field, and the very existence of this series of conferences exemplifies our collaborative desire to establish the problems of relevance and the appropriateness of methods.
Clearly, most people, and perhaps all people, exemplify these different types at different times and in different circumstances.
Schutz also examines how socially-derived knowledge originates and, again, uses ideal types to explore this phenomenon. Apart from ourselves, as the central (to our minds) figures in social interaction, he suggests that the sources of socially-acquired knowledge can be seen as four ideal types:
- the eyewitness, that is, someone who reports to me something that he or she has observed in the world within that person's reach;
- the insider: someone who, because of his relationship to a group which is more direct than my own, is able to report some event, or the opinions of others, with the authority of sharing the same system of relevance as the other members of the group. I accept the insider's information as 'true' or valid, at least in part, because the insider's knowledge of the context of the situation is deeper than my own;
- the analyst: someone who shares my system of relevances, who has collected information and organized that information in conformity with that system of relevance; and
- the commentator: someone who does not share my system of relevances, but who has collected information in the same way as the analyst and has presented that information in such a way that I can form 'a sufficiently clear and precise knowledge of the underlying deviating system of relevances' .
Again, each of us may, at different times and in different circumstances, take roles that fit these ideal types.
The essay on the 'well-informed citizen' demonstrates, I think, the power of the ideal type and, in this instance, points directly to the value that may be derived when thinking about information behaviour research. Schutz points out that ideal types are derived for a research purpose at hand and that the types may not be appropriate for a different research purpose. However, these types, derived as they are for the purposes of explicating the social distribution of knowledge, seem to me to be a very powerful tool for exploring information-seeking behaviour. A number of research questions seem to follow naturally from what Schutz has to say; for example:
- How does the 'man on the street' discover his recipe knowledge? What channels of communication does he use?
- Today, how does the existence of information resources on the Internet affect the way the man in the street thinks of finding recipe knowledge?
- What typical uses does the man in the street find for recipe knowledge?
- What are the boundaries, in terms of information-seeking behaviour, between representatives of the three ideal types?
- How does the 'man on the street' respond to the information provided by the eyewitness, the insider, the analyst and the commentator? What information is regarded as most 'relevant' or reliable, and why?
- Assuming that the scholarly record consists of information provided by 'analysts', under what circumstances might the scholar seek information from 'eyewitnesses', 'insiders', or 'commentators'?
Clearly, similar and many more questions could be asked in relation to these ideal types, but we have to recognize that to use ideal types in this way is not really phenomenological research: here we are using concepts derived from phenomenological analysis to evolve research questions to guide other modes of research.
Previous research in this field has led, of course, to other kinds of ideal types, although they are not always represented as such and the wealth of sociological connotation (in terms, for example, of systems of relevances) is usually ignored. For example, Palmer  identified ideal types of expert scientists as information seekers (using terms such as Hunters and Confident collectors in her study of behaviour in the Rothamsted Experimental Research Station. These are explicit ideal types, but the constructed searcher described by Ellis's  search characteristics, or by Kuhlthau's  stages, or by Wilson's  problem-solving process, also constitute ideal types (Schutz refers to this type of construct as 'puppets' created by the social scientist ). This fact is also generally ignored. Yet, when we postulate an ideal type on the basis of Ellis's characteristics, for example, it provides us with an opportunity to ask more research questions. We can enquire about the circumstances in which these characteristics are present, and when not, and, because they were derived from the behaviour of 'experts', we can also marry ideal types and ask whether and how these characteristics differ for the 'man in the street' and the 'well-informed citizen'.
With Schutz's concept of each ideal type being associated with different 'systems of relevances', we can also ask questions about the nature of those systems in each of those ideal types, or, indeed, about ideal types we may construct ourselves, if the concept 'system of relevances' is of relevance to our research.
FROM METHODOLOGY TO METHOD
There is a degree of confusion over the relationship between the concepts methodology and methods, we find, frequently, for example, that people write about 'methodology' when all they are doing is describing the choice of method for a study, or simply describing the method chosen. Methodology, however, is prior to method and more fundamental, it provides the philosophical groundwork for methods. To state one's methodological position is to describe one's view of the nature of reality: for the positivist, the methodological position is that the facts of the world represent real objects, while for the phenomenologist, the world (or at least the world s/he chooses to explore) is one of intersubjectively constructed meanings.
Within phenomenology, with its emphasis on understanding the person's experience of the world and his/her situation, the research methods are the methods of philosophy. Those methods include, for example, conceptual analysis; linguistic analysis; hermeneutical method and praxis; historical-critical method; literary philosophy; and formal logic. These methods also underlie some aspects of qualitative approaches in information science - conceptual analysis is at the base of all 'coding' activities in all methods for analysing qualitative data, and Cornelius  has suggested an hermeneutic approach to the whole of information science.
In phenomenological psychology we find the same kinds of research methods as in phenomenology as a philosophical discipline. The emphasis is on understanding the person's experience of the world and their situation and, therefore, narrative accounts, and qualitative interviews are regularly employed research methods. The mode of analysis will vary, of course, according to the theoretical perspective of the researcher, or, perhaps more appropriate to the phenomenological framework, according to the theoretical perspective that arises out of the data. In psychological research, for example, one may find, through extensive interviews with people suffering from the same diagnosed problem, that some experience the problem as something with which they have to deal in order to cope with the rest of their experience, whereas others find that the structure of their lives is determined by the problem - life is experienced through the problem. In other words, the concepts of structure, coping, 'problem life', and 'normal life', may emerge out of the analysis.
This suggests that methods are capable of being used under different methodological presuppositions, and this would be true. A questionnaire can be used to elicit different things and we can use the data obtained to satisfy our research objectives whether we adopt a positivist or a phenomenological stance. Elsewhere, I have set out a typology of research methods  that rejects the usual division into qualitative and quantitative methods, since I regard this division as confusing. For example, 'qualitative' interviews may be analysed to identify underlying concepts and the occurrence of these concepts in a succession of interviews may be counted and related to, for example, the demographic characteristics of the respondents, using SPSS for the analysis. Atlas.ti, the qualitative analysis package actually provides a means for converting one's analysis to SPSS input. Is the qualitative interview then a qualitative or a quantitative method?
My alternative typology suggests that observation is the fundamental method of data collection, and that this may be divided into direct observation by the researcher and 'indirect' observation, where the researcher collects self-observations from respondents. Either of these modes may have structure imposed on the data collection by the researcher, or s/he may allow structure to 'emerge' in the process of analysis. This classification results in the representation of methods shown in Figure 1.
Having suggested that research methods may be classified in this way, there remains the question, "What methods are appropriate to phenomenological research.", and here we must return to the fundamental methodological issues. Phenomenology demands that we seek to discover the world as it is experienced by those involved in it. It is about the nature of human experience and the meaning that people attach to their experiences. In trying to arrive at this kind of understanding, the researcher is asked to 'bracket', or suspend belief in the phenomena of the external world, to put them aside and focus on the consciousness of that world.
Consequently, phenomenology is not a hypothesis testing mode of research (although it may result in hypotheses that we may wish to test by other means), nor is it one that must be guided by theoretical models. Rather, one is urged to get as close as possible to what the participants in the behaviour of interest are experiencing. Clearly, then, this suggests that the research modes I have labelled as involving 'emergent structure' are most likely to be of use in a phenomenological investigation. In particular, observation and qualitative interviewing are widely used, as is documentary analysis, although this latter often goes under the heading of hermeneutics, which is closely related to phenomenology. In other words, the label 'qualitative methods', which my typology was intended to avoid, can be applied. In fact a well-known textbook, Bogdan and Taylor's Introduction to qualitative research methods  has the subtitle, A phenomenological approach to the social sciences.
Phenomenology is probably the most significant philosophical movement of the twentieth century, as far as the social sciences are concerned. Husserl's ideas, amended and developed, have informed research in sociology, psychology, social psychology, education, health sciences, and many other fields. Phenomenological ideas underlie virtually all of those schools of thought that hold that it is necessary to understand the meaning attributed by persons to the activities in which they engage, in order to understand their behaviour.
From this point of view, the move towards the adoption of qualitative methods in social research was prompted by the rise of phenomenological sociology, particularly through the social constructionism of Berger and Luckman  - students of Schutz. Other approaches in the social sciences, such as Blumer's social interactionism  - the source of Dervin's ideas on 'sense making' , have links to and much in common with phenomenology. For example, George Herbert Mead, Blumer's teacher, studied in Freiberg, and must have come into contact with the ideas that led Husserl to develop phenomenology as a philosophical discipline in the same institution. We see the connection, also, in Kuhlthau's use  of the Kelly's 'personal construct theory ' - Boeree suggests that this method is part of the armoury of the phenomenological investigator .
The intention of this paper has been twofold: to bring phenomenology into the general debate on theories of information behaviour, and to emphasise the distinction between methodology and method. Over recent years more attention has been devoted to research methods than to the underlying methodological principles of different research frameworks. I believe that Schutz's attempts to create a phenomenological sociology provide us with a useful framework to guide research into people's information behaviour and, thereby, to guide us to the choice of appropriate methods.
- HUSSERL, E. Ideas pertaining to a pure phenomenology and to a phenomenological philosophy. 1st book: General introduction to a pure phenomenology, trans. by F. Kersten. The Hague: Nijhoff, 1982
- DANIELS, V. Lecture on phenomenology. Rhonert Park, CA: Sonoma State University, 2000. Available at: http://www.sonoma.edu/people/daniels/phenomlect.html [Accessed 28th December 2001]
- WAGNER, H.R. Introduction. On phenomenology and social relations: selected writings, by Alfred Schutz. Edited by H. R. Wagner. Chicago: University of Chicago Press, 1970. 1-50
- WEBER, M. Economy and society: an outline of interpretive sociology. Vol.1. New York: Bedminster Press, 1968.
- SCHUTZ, A. Der sinnhafte Aufbau der sozialen Welt. Vienna: Springer, 1932
- SCHUTZ, A. The phenomenology of the social world. Evanston, IL: Northwestern University Press, 1967.
- SCHUTZ, A. Collected papers Vol. I. The problem of social reality. The Hague: Martinus Nijhoff, 1962.
- Ibid. p. 42
- SCHUTZ, A. The well-informed citizen. In: Collected papers. Vol. II. Studies in social theory. The Hague: Martinus Nijhoff, 1964, 120-134.
- MANNHEIM, K Ideology and utopia. London: Routledge & Kegan Paul, 1936.
- SCHUTZ, A. The stranger: an essay in social psychology. In: Collected papers. Vol. II. Studies in social theory. The Hague: Martinus Nijhoff, 1964, 93.
- SCHUTZ, A. Op. cit. The well-informed citizen. p.132
- PALMER, J. Scientists and information: I. Using cluster analysis to identify information style. Journal of Documentation, 47, 1991, 105-226.
- ELLIS, D. A behavioural approach to information retrieval design. Journal of Documentation, 46, 1989, 318-338.
- KUHLTHAU, C.C. Seeking meaning: a process approach to library and information services. Norwood, NJ: Ablex Publishing, 1993.
- WILSON, T.D. Exploring models of information behaviour: the 'Uncertainty' Project. In; T.D. Wilson and D.K. Allen, eds. Exploring the contexts of information behaviour: Proceedings of the 2nd International Conference on Information Seeking in Context, August 12-15, 1998. Sheffield, UK. London: Taylor Graham, 1999 55-66.
- SCHUTZ, A. Op. cit. Collected papers Vol. I. p.41
- CORNELIUS, I.V. Meaning and method in information studies. Norwood, N.J.: Ablex, 1996.
- WILSON, T.D. (2001) Mokslinio tyrimo metodu struktura: naujas ˛vilgsnis i tipologija. Knygotyra, 37, 248-252 [In Lithuanian: contact the author for an English version]
- BOGDAN, R. and TAYLOR, S.J. Introduction to qualitative research methods. New York, NY: Wiley, 1975
- BERGER P.L. and LUCKMAN, T. The social construction of reality :a treatise in the sociology of knowledge. Harmondsworth: Penguin, 1979.
- BLUMER, H. Symbolic interactionism: perspective and method. Englewood Cliffs: Prentice- Hall, 1969.
- DERVIN, B. An overview of sense-making research: concepts, methods, and results to date. Paper presented at the International Communication Association Annual Meeting, Dallas, Texas, USA, May l983. Available at http://communication.sbs.ohio-state.edu/sense-making/art/artdervin83.html [Site visited 11th February 2002]
- KUHLTHAU, C.C. Op. cit.
- KELLY, G.A. The psychology of personal constructs. New York, NY: Norton, 1955.
- BOEREE, C. G. Qualitative methods workbook. Available at: http://www.ship. edu/~cgboeree/qualmeth.html [Site visited 10th February 2002]
A paper delivered at ISIC4 - Fourth International Conference on Information Seeking in Context, Universidade Lusiada, Lisbon, Portugal, September 11 to 13, 2002