Part I: Comparative Orientations

Toward a Science of Comparative Education
On Teaching a 'Scientific' Comparative Education
Defining Comparative Education: Conceptions
A Comparative Study of Outlier Schools in Metropolitan Settings
Other Schools and Ours
Fast-Fish and Loose-Fish in Comparative Education
Use and Abuse of Comparative Education
State of the Field
Dependency Theory in Comparative Education
The Darling Young
The Comparative Mind: Metaphor in Comparative Education

Source: Harold J. Noah, "Fast-Fish and Loose-Fish in
Comparative Education," Comparative Education Review
18 (1974): 341-347.  Reprinted by permission of the
University of Chicago Press.

FAST-FISH AND LOOSE-FISH IN COMPARATIVE EDUCATION

A presidential address to a scholarly society is a fit moment for some reflections on the state of a field of study. Comparative and international education has grown considerably in the last two decades, as the membership of our Society, the contents of our journal, and the literature in our field all testify. We have come a long way since a small group met at New York University in the mid-1950s just 20 years ago under W. W. Brickman's informal chairmanship to promote a more permanent association of comparativists in education.

Yet, as we also know, this is a period of perceptible slowdown in the pace of international and area studies in general, and of comparative and international education in particular. It may be a convenient time to do a little general stock-taking, to assess what we have accomplished and what we know we still have to do.

Herman Melville provides me with the title and theme of my remarks. Along the way in Moby Dick, Melville tells us something about the law of the sea, as it pertains to the ownership of disputed whales. He entitles the chapter "Fast-Fish and Loose-Fish." Let me quote as briefly as Melville's masterly style allows me from that chapter:

"It frequently happens that when several ships are cruising in company, a whale may be struck by one vessel, then escape, and be finally killed and captured by another vessel.... For example, after a weary and perilous chase and capture of a whale, the body may get loose from the ship by reason of a violent storm; and drifting far away to leeward, be retaken by a second whaler, who, in a calm, snugly tows it alongside, without risk of life or line. Thus the most vexatious and violent disputes would often arise between the fishermen, were there not some written or unwritten, universal, undisputed law applicable to all cases....

"These laws might be engraven on a Queen Anne's farthing, or the barb of a Harpoon, and worn round the neck, so small are they.

"I. A Fast-Fish belongs to the party fast to it,

"II. A Loose-Fish is fair game for anybody who can soonest catch it.

"But what plays the mischief with this masterly code is the admirable brevity of it, which necessitates a vast volume of commentaries to expound it.

"First: What is a Fast-Fish? Alive or dead a fish is technically fast, when it is connected with an occupied ship or boat, by any medium at all controllable by the occupant or occupants -- a mast, an oar, a nine-inch cable, a telegraph wire, or a strand of cobweb, it is all the same.

Likewise a fish is technically fast when it bears a waif, or any other recognized symbol of possession; so long as the party waifing it plainly evince their ability at any time to take it alongside, as well as their intention so to do."

Possession is thus seen as the whole of the law, and Melville proceeds to extend the argument from whaling to social arrangements in general, nations and the world of ideas. He asks:

"What are the sinews and souls of Russian serfs and Republican slaves but FastFish, whereof possession is the whole of the law? What to the rapacious landlord is the widow's last mite but a Fast-Fish? What is yonder undetected villain's marble mansion with a door-plate for a waif; what is that but a Fast-Fish? ... What are the Duke of Dunder's hereditary towns and hamlets but Fast-Fish? ... And concerning all these, is not Possession the whole of the law?

"But if the doctrine of Fast-Fish be pretty generally applicable, the kindred doctrine of Loose-Fish is still more widely so. That is internationally and universally applicable.

"What was America in 1492 but a Loose-Fish, in which Columbus struck the Spanish standard by way of waifing it for his royal master and mistress? What was Poland to the Czar? What Greece to the Turk? What India to England? What at last will Mexico be to the United States? All Loose-Fish.

"What are the Rights of Man and the Liberties of the World but Loose-Fish? What all men's minds and opinions but Loose-Fish?

"What is the principle of religious belief in them but a Loose-Fish? What ... are the thoughts of thinkers but Loose-Fish? What is the great globe itself but a Loose-Fish! And what are you, reader, but a Loose-Fish and a Fast-Fish, too?"

And what, perhaps, is comparative education as an area of study but a FastFish and a Loose-Fish, too? Fast in the sense that we have made the comparative and international study of education our own area of specialization. A Loose-Fish, too, because many workers from other disciplines, finding that the phenomena of education and schooling provide a host of problems of exceptional interest, often seize on comparative analyses of education to advance understanding in their own fields.

In this address, then, I shall devote attention to the questions: What does our profession have a reasonably secure hold on? What, in a word, are our Fast-Fish? And, conversely, what are the major prizes out there, available for the taking, if we have the wit, energy and skill to catch them? What are the Loose-Fish for comparative education?

Our Fast-Fish

Year by year a great and growing body of information about the educational systems of the world has been accumulated. More and more, these national sources have been systematized and standardized so that they are better adapted to comparative work. For this work, we are particularly indebted to the international organizations, in particular to UNESCO (and especially to the Paris headquarters that has produced the successive editions of the World Surveys of Education and the Statistical Yearbook, as well as to the International Bureau of Education, Geneva), to the OECD, and to the World Bank. Although we can all think of improvements that might be made in these compilations, and there are still some serious and surprising gaps, they form a basic portion of our total comparative education catch, without which we would have been poor fishermen, indeed.

Second, the pages of our society's journal, the Comparative Education Review, and of other scholarly journals here and abroad report the progress we are making in understanding processes at work inside education and the connections between educational and more broadly social changes. Indeed, it is this analytic work that has most attracted the attention of and contributions from an ever-increasing number of scholars within education studies and from the other disciplines. Comparative studies of curricula, textbooks, classroom practices, teacher training and teachers' attitudes, students' aspirations, achievement outcomes, modes of educational finance, planning and administration, as well as the relationships between education and social status, incomes, wealth, cultural and political forms have all figured prominently in the expanding literature of the past decades.

Nor has all this been devoted to purely abstract theoretical understanding. A great deal of attention has been paid to the problems of reforming and restructuring educational practices to fit new needs, thus lending comparative studies an especially practical hue. Here, also, are to be found some of our Fast-Fish.

Perhaps the chief idea that we have advanced and defended successfully is the assertion that comparative education understanding is an essential part of teacher training, theoretical work in education, and the practical tasks of policy-making and administration in education. While in large and small countries alike the forces making for national self-sufficiency and nationalist concentration are as powerful as ever, at the same time there is increasing acceptance of the thesis that knowledge and appreciation of other nations' problems and solutions in education are indispensable. Not only is the nation that forgets its history likely to repeat it, but the nation that forgets (or is blind to) the educational systems of its contemporaries is risking either stagnation, or the perils of burdensomely expensive experimentation. Comparative understanding can help countries break with old ways of arranging the educational systems without the danger that they indulge in foolish daydreams that there are just one or two fairly simplistic things they need do in order to set their schools aright.

In this respect, special commendation is due to OECD for its series of Country Reports on Educational Policy, to the IIEP for its comparative studies of educational financial policies, and to several of the official commissions on educational reform in Britain in the past decade, who have been careful to include informative studies of other countries' experiences in their deliberations and final reports.

In all these efforts to establish some secure bases, some Fast-Fish, for comparative education, many approaches have been used, and this is as it should be. Some workers have chosen to emphasize that schooling in each country operates within a particular historically constrained context, and that an understanding of each context is the key to the understanding of each educational system, its success and its failures, its problems and its solutions, its strengths and its weaknesses. Others have been more concerned to abstract from the "givens" of each national system of education, in order to attempt statements of a more general kind about educational development. Some scholars rely heavily upon scaling, quantification, and relatively complex statistical techniques; others prefer to avoid the many problems associated with quantification and employ qualitative data. Studies run the gamut from world-wide, global analyses to single-nation case studies, with every variety and size of sample bracketed between the two extremes.

Each of these approaches has value for particular purposes in particular situations. None may claim a monopoly of truth and virtue. The task of the scholar and wise consumer of the fruits of scholarship is to recognize which approach gives the most useful results for a given purpose and in a given situation. To pursue Melville's analogy, there is no single way of making fast our fish. We need to select techniques well-suited to the conditions set by the problem, the data, the skills of the investigator, and the use to be made of the results.

The vital thing is to avoid conflicts over method in the abstract, but to ask of each piece of work using a comparative approach: Are the questions asked significant and clear? Are the data appropriate? In particular, do the data measure well the phenomena they purport to describe? Are they handled in a way that gives us confidence in the inferences drawn?

There are likely to be many routes leading to affirmative answers. In particular, the complementarity of qualitative and quantitative approaches needs to be stressed, as we work to make more secure our catch in comparative and international education.

Our Loose-Fish

Let me now turn to our Loose-Fish -- the catch that is not yet ours, but might be made so, if we are of a mind to engage in the chase. In what follows, I by no means intend to say or imply that members of our profession have not already begun the pursuit. Indeed, each of us knows of examples of the kind of work I shall mention. Rather, I should like to use this opportunity to underline the need, as I see it, for much more work along these lines.

Our field grew out of intense curiosity concerning different national systems of schooling. The great masters of the past Jullien, Arnold, Sadler, Kandel, Hans, and Rossello) and most of us today in our turn, view the nation state and the entire national educational systems as the "natural" units for comparative study. Inquiries into "the French educational system," "education in Iran," or "higher education in England" continue to form a major part of our concerns. We have been led, not unnaturally, to this position by some very obvious advantages of viewing the nation state as our primary frame of reference. Our data, whether quantitative or qualitative, have most often been collected according to these units; socio-educational and pedagogical policies are devised and administered nationally; national units provide the levers for decision and change; our lives are conventionally set in the perspective of our national allegiances.

But, in pursuing comparative studies in terms of national aggregates, I believe we have been inclined to neglect too much the other, complementary way of approaching our task: the cross-national comparison of sub-national units. For example, it seems desirable that we should have relatively more studies that deal cross-nationally with problems of learning and school achievement in specific social settings. Where are the comparative studies of "School and the Inner-City Child"?, or "Recruitment of Teachers for Rural Schools"?, or "Financing Schools in Suburban Areas"? We even lack comparative studies of school provision for the important minority groups that nearly every nation state contains. Perhaps what I am saying is that we need much more vigorous seizing of the avenue of inquiry opened by within-nation social science research, to test the applicability to a broader universe of conclusions reached within one nation. It is here, I would judge, that comparative and international studies of education can contribute most to the total catch of knowledge.

In line with these remarks is the notion that most of us tend to concentrate our studies too much at what I might call the macro end of the conceptual and data continua, and shortchange the micro investigation. For example, we have had a number of studies of educational reform in the last few years. They are heavily weighted toward the macro end of the data scale, dealing with general, large-institutional problems of promoting change. We have few, if any, comparative studies of the attitudes, opinions, and actions of samples of teachers, school officials, parents, children, and politicians in the matter of changing school structures and processes. We need to develop the human and financial resources to support research that will dig deeper under the surface of the aggregative, macro-institutional type of work that has typified comparative and international education to date.

There is a quite natural tendency for comparativists to assert that the fact that they deal with more than one nation, combined with limitations of time and research funds, means that they must be excused from getting too close to the micro-situations of particular children in particular classrooms, or particular teachers in particular schools. That, for example, instead of gaining knowledge about the attitudes and actions of small groups of educational bureaucrats in a number of countries, we must limit ourselves to the public utterances and debates, the legislation, the regulations, the formal generalized acts governing education on the national scale. While some studies of this kind are clearly desirable and in plentiful supply, we need to press more in the other direction, too. I suspect that there are some very interesting Loose-Fish out there to be made fast.

It occurs to me, too, that it is in this direction that our hopes for making comparative and international education useful to policy-makers and administrators really lie. Conclusions based on cross-national aggregative studies often make valuable background reading for decision-makers, but they too seldom illumine the actual decisions that have to be made. It is easy to see why this is so. Our aggregative studies normally proceed in terms of national average tendencies, with regard to both data and conclusions. National and local decision-makers in education, however, are typically concerned with the likely outcomes of taking particular steps in highly particular, rather than in average, circumstances. For example, to be able to say that, on average, across the entirety of most educational systems, higher levels of parental participation in school decisions are associated with higher rates of student retention is, no doubt, of general interest to educational administrators. But, if this kind of result is to be of practical use, it needs to be expressed in substantially more differentiated fashion. Does the assertion hold at all income levels? among minority as well as majority groups? in urban as well as in suburban, or rural areas? in non-state as much as in state schools? and so on. For it is on such differentiated bases, rather than by consideration of overall national aggregates, that most educational decisions and policies -- even the most important -- have to find both their political support and their final shape.

Two further aspects of our potential catch at first sight may appear to be merely technical, but they have in fact substantive implications. Perhaps the first point I shall make refers rather to the tools we use to catch our Loose-Fish (the nets, lines, harpoons, or what you will) than to the fish themselves.

There has been much discussion among ourselves, and in the social sciences in general, on the question of using quantified data, hypothesis testing and the usual array of statistical estimating techniques to try to make warranted statements about educational phenomena. As suggested earlier, my own view is that there is a role for many different kinds of work, and I am sure that work of a quantitative, statistical kind is here to stay and that our entire array of comparative studies in education will be the better for its presence.

But this is not to deny that serious weaknesses persist in the statistical work that has appeared. For me, perhaps, it is the explanatory models commonly used that pose the most serious and it may be the most intransigent difficulties. These models are correlational of one kind or another -- most often multiple regression, or path analysis. In such models the assumption is made that the effects of a complex patterning of variables on some outcome (say, school achievement) can be analyzed as if there could be attributed to each variable a constant strength of association between it and the dependent variable. This is, of course, the regression coefficient. Differences in outcomes in such models are explained fairly simplistically as the summed effects of different values of the independent variables, each multiplied by its own regression coefficient, plus an error term to capture errors in sampling, measurement and model specification. The linear regression model is a highly flexible approach. In particular, should the relationship between the independent and dependent variables be pretty obviously non-linear, there are usually available some quite simple techniques for transforming the data so that linear approximations may be used. Thus, the linear regression approach, in its various forms, has in its favor good explanatory power in multivariate situations, flexibility and generality of use, and hence an excellent standing among the statistical and social science fraternities. Why then do I call for something better?

The problem, as I see it, of applying these regression models to many of our explanatory problems in education (and, particularly when we try to explain variations in achievement, attitudes, career choices, dropout decisions, and the like) is that many of the processes we are trying to map (and certainly all of the pedagogical processes) are neither linear in their naturally occurring form, nor is it very helpful to force them into a framework where they can be handled as if they were linear. For, in reality pedagogical and learning processes are dominated by threshold effects, the presence or absence of catalysts, complex non-linear patterning, limiting factors, and the time-sequencing of events. If this is so, we should not be too surprised that even our most sophisticated multi- and cross-national work to date (I refer to the IEA studies) succeed in explaining relatively small fractions (for example, 30 to 40 percent) of total within-country and across-country variances in achievement. I urge that we accept the challenge to try to do better, by devising and applying more appropriate explanatory models.

Let me conclude this survey of some Loose-Fish I'd like to see us make fast by referring to a cognate statistical-cum-substantive outcome of conventional correlation and regression analyses. In such work, the investigator's attention is naturally focussed primarily upon the search for the regularities, generalizations, the "law-like" statements he can make. And this is quite proper. However, we may tend thereby to overlook those very interesting cases that may lie well off the regression plane, the outliers. Thus, the inner city school that produces excellent results in the face of grave measured disadvantages; the "good homes" that fail in using the conventional socio-economic paths to maintain and advance the class-status of their children; or the region that, against all expectation and statistical prediction, takes on to good effect an unusually heavy financial burden for the schooling of its children --these, and similar "outlier" situations may be the fattest Loose-Fish in our ocean of exploration. I commend them to your attention.

Our Society's Conference, this year as in past years, provides examples in abundance of the Fast-Fish and the Loose-Fish in our field. In some measure, I am sure, our activities over the forthcoming two days will assist us in reconfirming knowledge in areas we have already made reasonably fast; in others I hope it will help us to make what is somewhat loose a little more fast. Thank you for coming to Washington this year. And happy angling!