In its historical development geography has occupied a logically defensible position among the sciences as one of the chorographical studies, which, like the historical studies, attempt to consider not particular kinds of objects and phenomena in reality but actual sections of reality; which attempt to analyze and synthesize not processes of phenomena, but the associations of phenomena as related in sections of reality.

Whereas the historical studies consider temporal sections of reality, the chorographical studies consider spatial sections; geography, in particular, studies the spatial sections of the earth's surface, of the world. Geography is therefore true to its name; it studies the world, seeking to describe, and to interpret, the differences among its different parts, as seen at any one time, commonly the present time. This field it shares with no other branch of science; rather it brings together in this field parts of many other sciences. These parts, however, it does not merely add together in some convenient organization. The heterogeneous phenomena which these other sciences study by classes are not merely mixed together in terms of physical juxtaposition in the earth surface, but are causally interrelated in complex areal combinations. Geography must integrate the materials that other sciences study separately, in terms of the actual integrations which the heterogeneous phenomena form in different parts of the world. As Humboldt most effectively established, in practice as well as in theory, though any phenomenon studied in geography may at the same time be an object of study in some systematic field, geography is not an agglomeration of pieces of the systematic sciences: it integrates these phenomena according to its distinctive chorographic point of view.

Since geography cuts a section through all the systematic sciences, there is an intimate and mutual relation between it and each of those fields. On the one hand, geography takes from the systematic sciences all knowledge that it can effectively utilize in making its descriptions of phenomena and interpretations of their interrelations as accurate and certain as possible. This borrowed knowledge may include generic concepts or type classifications, developed in the systematic sciences; but, where these are found unsuitable for geographic purposes, geography must develop its own generic concepts and systems of classification.

In return, geography has contributed, and continues to contribute, much to the systematic sciences. In its naive examination of the interrelation of phenomena in the real world it discovers phenomena which the sophisticated academic view of the systematic sciences may not have observed, shows them to be worthy of study in themselves and thus adds to the field of the systematic studies. Further, geography constantly emphasizes one aspect of phenomena which is frequently lost sight of in the more theoretical approach of the systematic fields, namely, the geographic aspect. It serves, therefore, as a realistic critic whose function it is constantly to remind the systematic sciences that they cannot completely understand their phenomena by considering them only in terms of their common characteristics and processes. They must also note the differences in those phenomena that result from their actual location in different areas of the world. In order to interpret these differences correctly, and to interpret the resultant world distribution of their phenomena, the systematic sciences take from geography something of the particular techniques which its point of view has required it to develop--notably the techniques of maps and map interpretation.

Geography, like history, is essential to the full understanding of reality. The naked, schematic study of the systematic sciences divides up reality into academic compartments, and thereby necessarily destroys something of its essential character:

Geography adds, as Vidal said, "the aptitude of comprehending the correspondence and correlation of facts, be they in the terrestrial milieu which includes them all, be they in the regional milieu in which they are localized" [183, 299].

It is a corollary of this proposition that, in the application of science to society, as Finch observes, the chorological science of geography can function directly, since many of the problems of society--notably those concerned with the most efficient organization of land use--are, in fact, regional problems. But that statement does not mean--and I assume that Finch did not intend it to mean--that the chorological point of view requires the justification of utility [223, 21 ff.]. On the contrary, whatever value geography has in relating science to the problems of society, merely confirms the fact that in pure science itself--the pursuit of knowledge for the sake of gaining more knowledge--there is need for a science that interprets the realities of areal differentiation of the world as they are found, not only in terms of the differences in certain things from place to place, but also in terms of the total combination of phenomena in each place, different from those at every other place.

Geography, like history, is so comprehensive in character, that the ideally complete geographer, like the ideally complete historian, would have to know all about every science that has to do with the world, both of nature and of man. The converse of this proposition, however, is that every student of a systematic science is somewhat at home in some part of geography. Furthermore, both geography and history endeavor to describe and interpret actual sections of reality as they exist, and in these sections they observe phenomena by methods that, in a general way, are available to the common man. Consequently geography, like history, is a field apparently open for layman to enter. Whereas the study of history, other than current history, at least requires the degree of learning sufficient to utilize the records of the past, geography may be studied by any one who has the opportunity to travel and the ability to describe what he sees. Consequently, geography was in fact studied by laymen long before any organized subject of geography was constructed, and countless non-professional travelers since have contributed more or less useful data to its literature. This characteristic, likewise, it shares, for good or ill, with history.

In consequence, Richthofen bas noted, "many have the delusion that geography is a field in which one can reap without sowing. Because a great part of that which the serious research students have won in it is easily understood, one thinks that he can work successfully in it without preparatory training, and can win laurels by the easy means of describing fleeting travel observations or by uncritical compilations. An endless flood of superficial literature, which, in spite of its deficiencies, may not be denied the service of popularization, has been able to obscure the judgment of a great part even of the educated public concerning the scientific content of geography. But, just as with history, the apparent ease with which a great part of the facts secured can readily be understood, stands in contrast to the difficulty of sound research" [73, 68]. Allen Johnson, among many others, has discussed the importance of the same contrast in history [117].

Since the vulnerability of both geography and history to occasional trespass by wandering laymen is a result of the fundamental character of the field in each case, little would be gained by attempting to set up barbed wire fences in the form of erudite technical terms designed to bar trespass. Few geographers, presumably, will wish to have their subject strive for prestige by hiding its knowledge behind smoke-screens. On the contrary, in a subject in which the field includes vast areas that few professionals will have the opportunity to explore, the assistance of the interested amateur may heartily be welcomed. The sole provision that we might like to suggest is that the amateur, as in any activity of life, should recognize his need of securing as much knowledge and training from professionals as is possible for him, so that his efforts may produce results of greater accuracy and interest in themselves and of more lasting value for the science of geography.

Geography and history are alike in that they are integrating sciences concerned with studying the world. There is, therefore, a universal and mutual relation between them, even though their bases of integration are in a sense opposite--geography in terms of earth spaces, history in terms of periods of time. The interpretation of present geographic features requires some knowledge of their historical development; in this case history is the means to a geographic end. Likewise the interpretation of historical events requires some knowledge of their geographic background; in this case geography is the means to an historical end. Such combinations of the two opposite points of view are possible if the major emphasis is clearly and continuously maintained on one point of view. To combine them coordinately involves difficulties which, as yet at least, appear to be beyond the limitations of human thought. Possibly one approach to such a combination can be made in geography by the lantern-slide method of successive views of historical geographies of the same place. An attempt to develop a motion picture would produce a continuous variation with respect to both time and space which would, of course, represent reality in its completeness, but which appears to be beyond our capacity even to visualize, not to say, to interpret.

Though the point of view under which geography attempts to acquire knowledge of reality is distinct, the fundamental ideals which govern its pursuit of knowledge are the same as those of all parts of that total field of knowledge for which we have no other name than science.

Geography seeks to acquire a complete knowledge of the areal differentiation of the world, and therefore discriminates among the phenomena that vary in different parts of the world only in terms of their geographic significance--i.e., their relation to the total differentiation of areas. Phenomena significant to areal differentiation have areal expression--not necessarily in terms of physical extent over the ground, but as a characteristic of an area of more or less definite extent. Consequently, in studying the interrelation of these phenomena, geography depends first and fundamentally on the comparison of maps depicting the areal expression of individual phenomena, or of interrelated phenomena. In terms of scientific techniques, geography is represented in the world of knowledge primarily by its techniques of map use.

There are no set rules for determining which phenomena are, in general, of geographic significance. That must be determined, in any particular case, on the basis of the direct importance of the phenomenon to areal differentiation, and of its indirect importance through its causal relation to other phenomena. In order to determine his findings as accurately as possible, the individual student, in any particular case, must depend upon those among the significant phenomena for which he is able to secure some sort of measured data. Non-measurable, but geographically significant, phenomena must be studied indirectly, by whatever measurable effects they have produced.

These general principles lead to no general exclusion of any kind of phenomena, nor of any aspect of the field. In any particular study in systematic geography or in any partial study of a region, particular kinds of phenomena may logically be excluded only if they are not significant to the interrelations of those that are being studied. Finally, the ideal of completeness requires geography to consider not only those features and relationships that can be expressed in generic concepts but a great number of features and relationships that are essentially unique.

In order to make its knowledge of interrelated phenomena as accurate and as certain as possible, geography considers all kinds of facts involved in such relations and utilizes all possible means of determining the facts, so that results obtained from one set of facts, or by one method of observation, may be checked by those secured from other facts or from other observations.

With the same ends in view, geography accepts the universal scientific standards of precise logical reasoning based on specifically defined, if not standardized, concepts. It seeks to organize its field so that scholarly procedures of investigation and presentation may make possible, not an accumulation of unrelated fragments of individual evidence, but rather the organic growth of repeatedly checked and constantly reproductive research.

In order that the vast detail of the knowledge of the world may be simplified, geography seeks to establish generalized pictures of combinations of dissimilar parts of areas that will nevertheless be as nearly correct as the limitations of a generalization permit, and to establish generic concepts of common characteristics of phenomena, or phenomenon-complexes that shall describe with certainty the common characteristics that these features actually possess. On the basis of such generic concepts, geography seeks to establish principles of relationships between the phenomena that are areally related in the same or different areas, in order that it may correctly interpret the interrelations of such phenomena in any particular area.

Finally, geography seeks to organize its knowledge of the world into inter-connected systems, in order that any particular fragment of knowledge may be related to all others that bear upon it. The areal differentiation of the world involves the integration, for all points on the earth's surface, of the resultant of many interrelated, but in part independent, variables. The simultaneous integration all over the world of the resultant of all these variables cannot be organized into a single system. In systematic geography each particular element, or element-complex, that is geographically significant, is studied in terms of its relation to the total differentiation of areas, as it varies from place to place over the world, or any part of it. This is in no sense the complete study of that particular phenomenon, such as would be made in the appropriate systematic science, but the study of it solely in its geographic significance--namely in its own areal connections, and in the relations of its variations to those of other features that determine the character of areas. Although the study of any single earth feature is thus organized into a complete system in systematic geography, it is clear that at every point on the earth it is connected with the coordinate systems concerned with the other features.

In regional geography all the knowledge of the interrelations of all features at given places--obtained in part from the different systems of systematic geography--is integrated, in terms of the interrelations which those features have to each other, to provide the total geography of those places. The areal integration of an infinite number of place-integrations of factors varying somewhat independently in relation to place, is possible only by the arbitrary device of ignoring variations within small unit-areas so that these finite areal units, each arbitrarily distorted into a homogeneous unit, may be studied in their relations to each other as parts of larger areas. These larger areas are themselves but parts of still larger divisions--ultimately divisions of the world.

The problem of dividing the world, or any part of it, into subdivisions in which to focus the study of areas, is the most difficult problem of organization in regional geography. It is a task that involves a complete division of the world in a logical system, or systems, of division and subdivision, down to, ultimately, the approximately homogeneous units of areas. Difficult though the task may be, the principles of completeness and organization demand that geography seek the best possible solution.

One method of providing such an organization represents perhaps an intermediate step between systematic and regional geography. On the basis of any one element or element-complex--which latter may represent a great number and variety of closely related elements--we may construct a logical system of division and subdivision of the world according to types. Each of these systems of division determined on the basis of generic concepts of element-complexes, may be carried through by objective decisions based on measurement. Possibly as few as three such systems--each based on a cultural complex of many elements--may be adequate to provide outlines into which to organize most of our regional knowledge of the world. In each case, however, we are organizing separately different aspects of the geography of regions, we are not organizing the complete geography of regions.

A single system in which to organize the complete geography of the regions of the world must be based on the total character of areas, including their location as parts of larger units. Such a system of specific regions requires the consideration of all features significant in geography, some more significant in some areas, others in others. The determination of the divisions at any level involves, therefore, subjective judgment as to which features are more, which less, important in determining similarities and dissimilarities, and in determining the relative closeness of regional interrelations. At any level therefore, the regions are fragments of the land, so determined that we may most economically describe the character of each region,--that is, that in each region we will have a minimum number of different generalized descriptions of approximately similar units, each description involving the maximum number of nearly common characteristics and applicable to the maximum number of similar units.

Although all the fundamental ideals of science apply equally in all parts of geography, there are differences in the degree to which they can be attained in the different parts. These differences among the special divisions of geography--physical, economic, political, etc.--are differences in degree corresponding to the similar differences in degree to which the various systematic sciences are able to attain those ideals.

The greatest differences in character within geography are found between the two major methods of organizing geographic knowledge--systematic geography and regional geography--each of which includes its appropriate part of all the special fields. In addition to the difference in form of organization in the two parts, there is a radical difference in the extent to which knowledge may be expressed in universals, whether generic concepts or principles of relationships.

Systematic geography is organized in terms of particular phenomena of general geographic significance, each of which is studied in terms of the relations of its areal differentiation to that of the others. Its descriptive form is therefore similar to that of the systematic sciences. Like them, it seeks to establish generic concepts of the phenomena studied and universal principles of their relationships, but only in terms of significance to areal differentiation. No more than in the systematic sciences, however, can systematic geography hope to express all its knowledge in terms of universals; much must be expressed and studied as unique.

While there are no logical limitations to the development of generic concepts and principles in systematic geography, the nature of the phenomena and the relations between them that are studied in geography present many difficulties preventing the establishment of precise principles. These difficulties are of the same kind as are found, in differing degree, in all parts of science. In many of the systematic sciences, both natural and social, the degree of difficulty is as great, or greater, than in geography. In that field which is most nearly the counterpart of geography, namely history, the difficulties are in almost every case far greater. Systematic geography is therefore far more able to develop universals than is "systematic history." Nevertheless the degree of completeness, accuracy, and certainty, both of the principles established and of the facts known in regard to any particular situation, seldom permit definite predictions in geography. This characteristic, geography shares not only with history, but also with many other sciences, both natural and social.

Regional geography organizes the knowledge of all interrelated forms of areal differentiation in individual units of area, which it must organize into a system of division and subdivision of the total earth surface. Its form of description involves two steps. It must first express, by analysis and synthesis, the integration of all interrelated features at individual unit places, and must then express, by analysis and synthesis, the integration of all such unit places within a given area. In order to make this possible, it must distort reality to the extent of considering small but finite areas as homogeneous units which can be compared with each other and added together in areal patterns of larger units. These larger, likewise arbitrary units, are so determined as to make possible a minimum of generalized description of each unit "region," that will involve a minimum of inaccuracy and incompleteness.

Since the units with which it deals are neither real phenomena nor real units but, at any level of division, represent distortions of reality, regional geography itself cannot develop either generic concepts or principles of reality. For the interpretation of its findings it depends upon generic concepts and principles developed in systematic geography. Furthermore, by comparing different units of area that are in part similar, it can test and correct the universals developed in systematic geography.

The direct subject of regional geography is the uniquely varying character of the earth's surface--a single unit which can only be divided arbitrarily into parts that, at any level of division, are, like the temporal parts of history, unique in total character. Consequently the findings of regional geography, though they include interpretations of details, are in large part descriptive. The discovery, analysis and synthesis of the unique is not to be dismissed as "mere description"; on the contrary, it represents an essential function of science, and the only function that it can perform in studying the unique. To know and understand fully the character of the unique is to know it completely; no universals need be evolved, other than the general law of geography that all its areas are unique.

In the same way that science as a whole requires both the systematic fields that study particular kinds of phenomena and the integrating fields that study the ways in which those phenomena are actually related as they are found in reality, so geography requires both its systematic and its regional methods of study of phenomena and organization of knowledge. Systematic geography is essential to an understanding of the areal differences in each kind of phenomena and the principles governing their relations to each other. This alone, however, cannot provide a comprehension of the individual earth units, but rather divests them of the fullness of their color and life. To comprehend the full character of each area in comparison with others, we must examine the totality of related features as that is found in different units of area--i.e., regional geography. Though each of these methods represents a different point of view, both are essential to the single purpose of geography and therefore are properly included in the unified field. Further, the two methods are intimately related and essential to each other. The ultimate purpose of geography, the study of areal differentiation of the world, is most dearly expressed in regional geography; only by constantly maintaining its relation to regional geography can systematic geography hold to the purpose of geography and not disappear into other sciences. On the other hand, regional geography in itself is sterile; without the continuous fertilization of generic concepts and principles from systematic geography, it could not advance to higher degrees of accuracy and certainty in interpretation of its findings.

It would be an error to interpret the current interest in methodological discussion as a sign that American geography had entered a period of unusual dissension. In large part, no doubt, it represents the crystallization in print of disagreements hitherto held in the more liquid solution of oral discussion, but which have been suddenly precipitated by the first challenge to fundamental principles in geography to appear in print in more than a decade. Although the basic position of geography as a chorographic science has been questioned, the challenge does not appear to have produced dissension. On the contrary, it has revealed that those who were accustomed to find themselves in opposite camps in methodological discussions, were actually in opposition only on secondary questions and were fundamentally at one on the major function of geography among the sciences.

More than at any previous time in the development of American geography, there is notable agreement, in practice as well as in theory, on the importance of studies in regional geography, while at the same time there is a continued drive to develop the various aspects of systematic geography. Further, the apparent gulf between these two aspects of the field is being narrowed: students of regional geography depend increasingly on studies in systematic geography and those making systematic studies have recognized that their value to geography as a whole depends on the extent to which they are correlated with the viewpoint of regional geography.

If American geography is approaching that major degree of common understanding on the fundamental nature of its field that was attained in Germany two or three decades earlier, and likewise underlies--even though less definitely expressed--a great part of the work in French geography, we may hope that the immediate future in this country will show a period of correspondingly rich production along a wide, but common front. Agreement on methodological questions, as on any others, can be attained, by those who are free to think for themselves, only by thorough examination of the problems involved, with adequate and fair consideration of the divergent views expressed by other students, past as well as present. By directing on current methodological problems in our field a critical review organized out of the rich literature of more than a century of geographic thought, I hope to have contributed to a more general understanding of our fundamental purposes and problems.

University of Minnesota.
June, 1939.