Questions regarding the role of the Ph.D. degree in our professional field arise, in my experience, quite frequently and in a variety of environments. Among my academic colleagues at the university, there seems to be an unwritten assumption, which sometimes emerges in very explicit form, that the letters and sciences are not only the heart of the university, but are also somehow on a higher plane than professional fields. Research in those areas is "pure," in professions "applied," and, therefore, lesser.

On the other hand, among at least some library practitioners, one can find a quite pronounced view that we who are teaching in professional programs are foggy-headed researchers—hopelessly impractical, incapable of preparing our students for the real world of practice. More specifically, our research is often seen to be so unconnected to real-world situations that it can offer little of use to the working professional.

So what is it that we are doing when we get a doctorate in this profession? Doing "merely" applied research, which our academic colleagues dismiss, while, on the other hand, the practitioners view as so impractical that it is of no use anyway?

On the positive side, almost all universities recognize that researchers in professional fields can earn the Ph.D., that is, a doctor of philosophy in a profession. In other words, they recognize that even professions can do research that taps basic questions, and are not only applied fields. By the way, in our field, there have only been one or two cases that I know of where a School has both a DLS, or Doctor of Library Science, and a Ph.D. In most places, even where the student earns a DLS, such as at the now-closed school at Columbia University, the degree is understood to be a Ph.D., de facto. Only where both degrees exist is a distinction is made between them.

So just what is the nature of a Ph.D. in a professional field—specifically, in this one? Let’s look first at what a profession is. A profession develops a body of knowledge around roles and activities that a society values. Professional skill then involves mastering that body of knowledge and developing experience applying it to situations where the knowledge is relevant.

, I d

Professional knowledge is not algorithmic; that is, it does not involve following a rule-based sequence of actions that are applicable in every common situation. Rather, a profession always requires developed judgment. There are complexities, subtleties, and variations from situation to situation that require experienced judgment in order to apply the professional knowledge properly.

On the other hand, the knowledge a profession develops always has some application to real situations in life. Knowledge developed in a research discipline may be abstruse and complex, and certainly not algorithmic, but if it has no application to some identifiable real situations, then it is, in the time-worn phrase, "purely academic," and not professional knowledge.

Professions develop out of social need and out of social consciousness. In the nineteenth century, when the size of library collections reached a point where older ad hoc and idiosyncratic approaches would no longer suffice, standards, training, and a self-conscious world view toward library and information services began developing. At the same time, a conceptualization of the very idea of a profession was also ripening and maturing. Medicine, teaching, nursing, the law had all been around for centuries, but their conscious formulation as professions is a fairly recent development.

Indeed, there are many social forces shaping the modern idea of a profession, and our profession in particular. But today, I want to concentrate on the question I started with--what is the role of the Ph.D. in a professional field? And I want to address that question with respect to the intellectual content of Ph.D.-level research and its relation to the profession of library and information science--or, as it is becoming known, information studies.

To cut to the chase, the fundamental point I want to make is that doctoral level research is the seed-bed for ideas and practices of the profession. AND it achieves that role by doing research that is as fundamental as any done in the letters and sciences.

I have argued elsewhere that information studies is fundamentally socio-technical. Our professional functions are derived from social needs and circumstances, and, to be successful, must serve to meet those needs. Information studies meets those social needs by drawing on a wide range of intellectual and physical technologies. Achieving the objectives of our profession depends on the skillful melding of an understanding of social, political, economic, and psychological factors related to information needs and uses with a mastery of both theoretical and practical issues in information organization and system design.

There is something of a culture war going on these days between those in the field who see the institution of the library and all it has meant to our society as the defining core of our field and those who employ the concept of information to define the field. To the "library" people, the information perspective is coldly technological and indifferent to important social values and a rich history of service to a wide range of people, many of whom were socially and economically enfranchised by the education they received at the library. To these "library" partisans, the information perspective means pure computer technology, carrying with it implicit authoritarian, even fascist, adoration of the power of technology, to the detriment of precious human values.

I confess that I nonetheless come down on the "information" side of this argument, and that, needless to say, I have a different perspective on the information view of our field. Though I do not believe that most "information" partisans carry such an adoring view of technology, to whatever extent they do, I side with the "library" partisans. Librarianship has precious values of service to society, and I would like to see those values carry over to any conceptualization of our field as information studies.

However o think the "information studies" approach is the right one for now in our field. To many people, the term "information" is a cold and neutral one, and is limited in their minds to pieces of data, like the population of Turkey or the distance to the sun. In this field, however, we can and should use the term much more broadly, and apply it to all experiences that can or do impact the human understanding and memory. Such a concept is indisputably rooted then not only in the single mind, but also, necessarily, in human society, as the mind is never wholly independent of the social matrix in which it was nurtured and continues to act daily.

Some would then answer, well, in that case, "information" means everything--all knowledge and experience. No one field can possibly encompass all of that. I agree, and I don't make such a claim. We do address all of knowledge, but in a very particular way. Information studies is a meta-field, like education or communication. Those fields, likewise, deal with all knowledge, but each does something distinctive with that world of knowledge and experience that is unique to each field.

Our field organizes information systems and services to offer that knowledge and experience to information seekers, education shapes curricula and uses teaching skills to convey that knowledge to learners, and communication/journalism uses reporting and writing skills to convey discovered news to others. Each field holds a particular rhetorical stance toward all knowledge, a stance that enables it to understand the human relationship to information in its own distinctive way, and in service to a larger social purpose that is unique to each field.

Once we see information studies in this manner, it becomes clear that the activities traditionally engaged in by librarians in libraries bear many similarities to the activities of archivists, of professionals in technical documentation, paper and electronic records, digital databases and libraries, and so on. The shift to information studies is not, should not, involve any rejection of libraries or librarianship; rather it should broaden our perspective to all the contexts within which these meta-field activities take place.

However this debate in the field over our fundamental world view and philosophy is resolved, the fact that there are such debates in the first place should provide a clue to the depth of thinking and theory that lies at the heart of our work. In truly algorithmic work, such as that of a restaurant cashier, it is doubtful that there are raging debates about the philosophy of cashiering.

I would like now to provide an extended example of how our work draws upon, and needs to draw upon, research that is as fundamental as any in the letters and sciences. The example involves the design of information retrieval systems. Now, surely this is at the technical, applied end of the business if anything is, right? Well, let's see.

Over the last two or three years I have come to see that the design of information systems is vastly more complex than it was in the early days of automated information systems. Early online database systems were crude and clunky. We didn't mind, however, because it was so exciting that we could search online for information in real time. People mastered arcane techniques and worked around the limitations of these systems as they needed to.

By now, however, the vast storage capabilities, sophisticated computer programming languages, the varieties of search engines, the speed of networks, and the subtlety of database design have all opened up a vast array of new design possibilities and complexities. Furthermore, these technical elements I've mentioned by no means exhaust the design elements that must be taken into consideration in creating the contemporary information system. The selection of the information content itself, of systems of metadata and indexing, design of search capabilities, selection or design of various kinds of access front-ends, as well as interface design, must all go into the design of an automated information system.

However, there remain still two more huge factors in this system design that are essential to its success. First, the needs and characteristics of anticipated users of the system must be taken into account with respect to every decision made on each of these technical and informational elements. Second, all the design elements or layers need to be taken into account simultaneously and in relation to each other.

Still, however, despite the complexity, all this might still be viewed by some as a merely technical or applied problem--inviting the conventional dismissal. Now, in fact, design of all kinds, including information system design, requires as much creativity and originality as anything—but that’s another speech. Here, I want to address the more conventional, recognized forms of academic research.

So let us look deeper. To make the following points, I am drawing on an extensive study done by the Getty Information Institute, which I participated in and wrote six articles about. In the study, Visiting Scholars in the humanities at the Getty were given the opportunity to be trained to search databases provided by the vendor DIALOG, and then do as much searching as they wished, for free, during their year as Visiting Scholars. In exchange, they would participate in interviews and agree to have their entire searches recorded automatically by computer.

In such a case, when we examine the users' needs, that examination must be conducted as the very subtlest of original, basic social science research. To design an information system that people will actually use, we must understand, in a profound way, how people use and relate to the information in their domains of interest. In this case, we were dealing with humanities scholars who had had, sometimes, decades in which to develop research styles and in which to shape their perspectives on the world of recorded information that they dealt with so intensively as scholars.

Here is where the term "information" must be used in the broader and deeper sense I was arguing for earlier. In particular, in the Getty study it became clear to me that we could not understand how the scholars used their searching opportunity, and could not suggest features needed for them in information systems, without understanding something of the entire social and educational system of practices and values within which they operated. This is not simply a matter of gathering statistics on what percentage of the scholars’ resources were books, what percentage journals, and the like.

Rather, it involved developing a deep and subtle understanding of the culture of humanities scholarship. It was necessary to understand how the scholars conceptualized their research intellectually, and how they conceptualized the process of going about locating and using research materials. It became clear that many assumptions in information science about methods of research and desirable system design characteristics were rooted in the research practices of the sciences and were inapplicable to the humanities.

For example:

• We learned that the selection of resources for the typical database—with a heavy emphasis on current journal articles—was much less useful for the scholars than for scientists. Currency is very important in the sciences, while historical range is more important in the humanities. This may seem obvious, but this basic reality was seldom understood and acted upon by the database producers, who were creating products on the science model.

• Second, fundamental differences in the way scholars conduct research and in how they attempt to find new information require at least a partial reconceptualization of information retrieval. Again, the paradigm of information retrieval research has been based on the science model. Some of the basics of that model do not make sense for the humanities scholar.

One such assumption is that the user of an information system is unfamiliar with the majority of the records in the system relating to their topic of interest. We found, however, for reasons that are based in the way scholars see their own relationship to their literature, compared to the way that scientists do, that humanities scholars almost always knew most of the literature in their field already. Thus, they saw their task as scholars to be to achieve mastery of a large number of particular works by other scholars.

For scientists, on the other hand, knowledge of particular publications is of less interest than knowledge of work going on in other laboratories. Scientific researchers on the cutting edge will know of discoveries long before they reach print in any form—thus the print version generally serves only an archival function.

In humanities scholarship, the particular insights achieved by the scholar are so important that the texts of publications must be read closely and absorbed, then reread. I am generalizing, of course, but there were nonetheless marked trends that can be seen to distinguish the two groups of researchers. Most of the time, for scholars, a database search functioned only as a way to catch any stray materials they were not already familiar with. If such is the typical situation of the humanities information system user, then we may need to design information systems in ingenious new ways that the science model did not suggest. The fourth and fifth reports in the Getty series explain these points in more detail, and suggest further design implications.

• Third, we found that the nature of humanities queries differed from those in the sciences not only in the obvious respect of content, but also in conceptual structure. I have argued that one consequence of this systematic difference is that humanities materials should probably be indexed by faceted vocabularies, not by conventional alphabetical thesauri in the scientific model. Furthermore, given this presumptive mode of indexing, then information system search capabilities and interface characteristics should be designed quite differently to accommodate these inherent information and system differences.

In fact, virtually every layer in the information system design would be affected dramatically if these fundamental social characteristics of humanities research and researchers were taken into account.

I’ve described social factors that go very deep into the information seeking and using behavior of humanities scholars. Along the same lines, we could describe ergonomic, psychological, and perceptual characteristics of people that must be taken into account for system design, based on basic research in the cognitive sciences. Further, we could identify—and this is unique to the information sciences—social, statistical, and economic characteristics of the humanities literatures themselves that need to be understood to optimize information system design.

If we think of the natural sciences as studying the physical world as their universe, the social sciences studying the social world, and the arts and humanities studying the creative products of human beings, then we can think of the information sciences as taking the universe of recorded information—and human beings’ interactions with it—as our domain of study. It is uniquely our challenge to identify the statistical, social, economic, and design properties of the world of recorded information.

At every point, research results of basic research must be drawn upon to create optimal information systems. Frequently, the only people doing the kind of basic research we need in this field, are people in the field.

I could make analogous arguments about the need for basic research as a grounding to developing information policies, or to developing the best institutional bases for information services. In other words, in every area, this profession must draw on original, basic research to achieve the best professional service.

This brings us to the other problem I mentioned earlier, however. If those of us with Ph.D.’s in information studies are busy doing basic research, or developing experimental systems and services that draw on basic research, how does this effort affect the working practitioner? There have been many lamentations, at conferences and in print, that researchers keep doing things that the practitioner cannot directly make use of. Is the availability of the Ph.D. degree creating a group of irrelevant people producing results no one will use?

The first thing we need to understand in relation to this question is that this is a near-universal problem, not one specific to our field. It is a problem in the letters and sciences and a problem in the professions. The bench chemist in a pharmaceutical firm no more uses the latest chemistry results, as a rule, than the working librarian applies the latest results published in Library Quarterly or the Journal of the American Society for Information Science.

The problem arises for reasons very similar to the ones mentioned earlier regarding the differences between scholars and scientists. Scholars and scientists have different cultures, values, and ways of operating because the nature of their work demands it. The same holds true for researchers and practitioners. Different native talents, training, and experience all add up to create people—professors and practitioners--who work and produce differently, and have different cultures and outlooks. Distinct types of minds and personalities will be drawn to these distinct activities.

I have spent just two years as a manager—as the Chair of my Department. And at the end of the two years I was a walking nervous breakdown. I have boundless admiration for Dean Robbins and other academic leaders who skillfully navigate the dangerous waters of large egos, small budgets, and killer work schedules. People who have a good understanding of research, and who are simultaneously good at the practice of academic management are rare, and universities are perennially, and often vainly, hunting for them.

It is in the nature of basic research that it often does not have immediate applications. But even where there are immediate applications, the work produced by the researcher often cannot be translated readily by the practitioner into practice. At this point the criticisms arise.

It needs to be understood by all involved that the ability to translate between different intellectual and practice cultures is a very special talent, and not all that common in any field. Most researchers cannot do it and most practitioners cannot. Those who can, should be treasured.

Unfortunately, the rewards are few to the people who engage in this translation process. The librarian is unlikely to have the time to devote to the extensive literature review and analysis needed to prepare a research-into-practice article—and may not have the research training to understand all the literature. The researcher, on the other hand, gets scant recognition or merit raises for producing what are generally dismissed as "service publications." Consequently, those of us in academia seldom write such boundary-spanning works.

To sum up, I believe that the Ph.D. in a profession is viewed as impure or applied precisely because it must hold a broader view of its subject matter. The holder of a Ph.D. in a professional field must be able to do basic research AND select research questions that may contribute—if not this century then the next—to the practice of that socially useful activity that is a profession. I’m hinting here that doing good research in the professions may actually require more talent than the work in the traditional disciplines. At the very least, it requires a considerable breadth of vision, a flexibility, and an ability to see connections others may miss.

Few are the researchers who can take it still a step further and extract implications for practice, and present those implications in a way that practitioners can make immediate use of. That is quite a challenge. Where that ability does exist, however, it deserves to be richly rewarded.