Chapter I, Historical Understanding, by Louis O.Mink, ed. Brian Fay, Eugene O. Golob, and Richard T. Vann. Ithaca and London: Cornell University Press, 1987, pp. 35-41. From Proceedings of the Xllth International Congress of Philosophy (Firenze: Sansoni Editore, 1960) V, 41 1-17 .
Modes of Comprehension and the Unity of Knowledge
Attempts to construct a general theory of knowledge--in fact, every attempt at a philosophical synthesis since Democritus--have taken as a model of knowledge some area of positive science and extended its procedures, presuppositions, and purposes to a definition of knowledge as such. For philosophical rationalism before the nineteenth century, the model was always mathematics. For Hegel and for Croce, the model was historical knowledge. Modern empiricism has tended to take theoretical physics as its model, and most recently some English-speaking philosophers have been moving in the direction of regarding legal argument as exemplifying the normal type of thinking. There have of course also been irenic attempts to settle the strife of models by distributing the field of knowledge among several sovereign sciences, for example by distinguishing between the Naturwissenschaffen and the Geisteswissenschaften or by Windelband's distinction between "idiographic" (i.e., particularizing) and "nomothetic" (i.e., theoretical) sciences. Such distinctions in effect abandon the great historic ideal of the unity of knowledge either by recognizing an irreconcilable plurality of methods or by a division of the world into independent subject matters. But it could not be said that any such proposal has been enduringly successful, or could be. The irresistible tendency of any general method is imperialistic; it inescapably prescribes its own subject matter and rejects as irrelevant and unreal whatever cannot be brought under its hegemony.
But the great debate between proposals of method turns on differences of analysis of the sufficient conditions of knowledge, and in the radical disagreements about sufficient conditions it has been easy to overlook the remarkable consensus which exists about some of the necessary conditions of knowledge. It is the purpose of this essay to call attention to one such neglected condition and to show that a consideration of it suggests a new question about the unity of knowledge. It appears that the separate sovereignties on the map of knowledge must be distinguished not by methods nor by subject matters but by unique and irreducible modes of comprehending the world as a totality.
The fact to which any theory of knowledge must return is the simple fact that experiences come to us seriatim in time and yet must be capable of being held together in an image of the manifold of events. The steps of a proof, the actions of a narrative, the notes of a melody, and even the words of a sentence are experienced one after the other, but must be considered in a single mental act before they even constitute data for significant discourse. Such an act, which may be called "comprehension," differs from both judgment and inference and is in fact presupposed by both. The assertion or denial of a relation between concepts, which is the characteristic of judgment, presupposes the act of considering the concepts together; and the scope of comprehension is wider in any case than the domain of concepts. It is represented in the act by which one thinks of a sonnet as a whole, distinguished from the line-by-line reading of a sonnet for the first time, or in the act by which one thinks of a historical event in a context of other events. Moreover, where inference refers to the drawing of a conclusion from premises, comprehension refers to the capacity to think of the conclusion together with the premises, as in mathematical demonstrations it is not uncommon to grasp a proof as a whole rather than as a series of manipulations according to rule.
As a phenomenon, comprehension is indeed so ubiquitous that, like other constant features of experience, it has been overlooked in favor of more variable and vivid data. But it has also been frequently recognized even when not named and described. Descartes seems to have had it in mind in stating his fourth rule of method in the Discourse: "to make enumerations so complete and reviews so general that I should be certain of having omitted nothing." If this rule is merely a reminder to check one's work as one would proofread a manuscript or check the addition of a column of figures, it is no doubt  salutary advice but unworthy of inclusion in the four rules which Descartes proposed to substitute for the "great number of precepts of which Logic is composed." But the rule deserves its place if it is interpreted as indicating the synthesis toward which analysis is directed, that is, as a practical precept for arriving at that state of comprehension in which the great number of elements into which Cartesian analysis has divided a problem are grasped together in a single mental act. Descartes says as much when he explains the process by which he discovered analytic geometry (Discourse, Part II): "In order to keep them [geometrical relationships and proportions] in my memory or to embrace several at once, it would be essential that I should explain them by means of certain formulae, the shorter the better."
Plato himself regarded comprehension as characterizing the final wisdom resulting from inquiry; it is the point in the education of the Guardians in the Republic at which, following their mastery of the individual mathematical sciences, "reflection can take a comprehensive view of the mutual relations and affinities which bind all these sciences together" (531 D). Now of course the method by which Plato proposed to attain a "comprehensive view" differs from Descartes's method, and the elements and relations comprised by comprehension differ as well in the two cases, but it is at least notable that they agree in regarding comprehension as a kind of totum simul, the grasping in a single act as a totality of what the discursive intellect otherwise can review only seriatim. Nor have rationalists alone implicitly or explicitly recognized comprehension. Empiricists, in their emphasis on inductions from particular cases, require it too as the condition of considering together the multiple cases on which any particular induction rests. Bergson, almost alone in recent philosophy, has made comprehension the cardinal principle of a philosophical theory, rather than its presupposition and ideal aim, but his error lay just in this, that he regarded as a unique method--which he called "intuition"—what is in fact the aim of every method. Hence he regarded intuition and analysis as opposed and incompatible methods, whereas comprehension, which phenomenologically seems like intuition, is not a method but the grasp of totality without which any method would disappear into the stream of unreflective consciousness.
Independent as such of particular method, comprehension operates at all levels of reflection and inquiry. At the lowest level, it is a  grasping of the data of experience and issues in the perception and recognition of objects. At an intermediate level, it is the classification together of a set of objects and issues in the formation of concepts. At the highest level, it is the attempt to order our knowledge of the world into a single object of understanding. Of course this is an unattainable goal, but significant nonetheless as delineating the teal by which partial comprehension is measured. Naturally enough, one finds the goal described in theological or quasi-theological terms. Laplace's omniscient scientist, for example, knowing the laws of nature and the position and velocity of every particle in the universe at an instant, could predict and retrodict the detailed character of the world at any moment of time. Boethius, in a different image of totality, described God's knowledge as a totum simul in which all moments of time would be seen as simultaneously present in a single divine perception--history spread out in a single panorama as a landscape is for us. And Plato called divine the knowledge which would consist in the contemplative vision of a set of essences apprehended as a single intelligible system. As much as these three may differ as theories of knowledge, they agree in characterizing the state of knowledge as a complex but unitary act of mind.
Yet at the same time these three instances suggest that there are fundamentally different modes of comprehension. A number of objects may be comprehended, as for Laplace and for Descartes, as instances of the same generalization or formula or law. This is a very powerful but very thin sort of comprehension: it is powerful because the generalization refers to its objects as members of a class and is a way of comprising them all, experienced and unexperienced, actual and possible. But it is thin because it refers to these objects only in virtue of :heir possession of certain common characteristics and omits everything else in the individuality of each. In general, one may, through such hypothetico-deductive or theoretical comprehension, understand all the instances which are consequences of a hypothesis. A physical Law orders all the phenomena which exemplify it, as Boyle's law explains both volcanoes and steam engines; similarly, a formal logical or mathematical system is entirely understood when the postulate set, the definitions, and the rules of inference are known. No doubt the requirement of economy, which is Occam's razor applied to formal systems, reflects the fact that a simpler postulate set is easier to comprehend as a whole than a more cumbersome one which might be equally useful from the standpoint of technical application. 
But there is another way in which a number of things may be comprehended, as elements in a single complex of concrete relationships. It is in this way that we see together the multiple images and allusions of a poem, or the combination of influences, motives, beliefs, and purposes which explain a concrete historical action. It is not as instances of a theory but as centers of concrete relationships that we understand ourselves and others, and one may say that there is also a kind of configurational comprehension. As the theoretical mode of comprehension corresponds to what Pascal called l'esprit de geometrie, so the configurational mode corresponds to what he called l'esprit de finesse, the ability to hold together a number of elements in nice balance.
This is the mode of comprehension adumbrated by Boethius. Yet there is a third way, envisioned by Plato: to hold together a number of things as examples of the same category, and in fact of a system of categories incapable of abstraction from each other. If this were identical with theoretical comprehension, philosophy would indeed be a rigorous science, but indeed only one among many. But there are many reasons why subsumption under categories is not identical with deduction from hypotheses. One difference is that hypotheses can be given meaning independently of each other (even though the truth of a hypothesis may be logically connected with the truth of other hypotheses), whereas the meaning of categories is essentially dependent on their systematic interconnection. Thus categoreal comprehension is neither so powerful nor so thin as theoretical comprehension. Moreover, categoreal connections, unlike theoretical hypotheses, are not falsifiable by particular experiences, because they give form to experience itself (in this Kant was clearly right). Speaking roughly, one might say that theoretical comprehension emphasizes the relations that may hold between universals and particulars, configurational comprehension the relations that may hold between particulars and particulars, and categoreal comprehension the relations that may hold between universals and universals. Subject as this formulation is to correction and expansion, it serves to indicate that these three modes exhaust the possibilities.
It should by now be apparent that theoretical, configurational, and categoreal comprehension are respectively appropriate to the natural sciences, to history, and to philosophy. But not entirely. Despite loose affinities, it would be vain to argue that these modes are coextensive with disciplines whose limits have been set by accidents of  history. Moreover, there have been and will no doubt continue to be attempts and proposals to reorganize, for example, historical inquiry with theoretical comprehension as its aim, or philosophy with configurational comprehension as its aim. But one cannot argue for the primacy of any of the modes except by reference to criteria which themselves are derivative from that mode's aim of comprehension. Hence each mode is self-justifying; critical analysis and intellectual advance are possible within but only within each mode. In each case the aim of ultimate comprehension leaves open the question of which theories, configurations, or category-systems will prove satisfactory by the standards relevant to the aim. Thus while each from its own standpoint envisions a unity of knowledge, and regards the others as errors or as subordinate stages in its own development, one must conclude that they constitute irreducible perspectives.
Each mode, in fact, has the totality of human experience as its subject matter. A mountain range is one sort of fact for the geologist, another for the historian, and yet another for the aesthetician, although this difference to their cognitive aims may be obscured because the mountain range has the same relation to their practical interest as, for example, travelers. It would, of course, be a mistake to try to understand subatomic particles in any mode other than the theoretical; but this is because such particles are not data of experience and very possibly not facts of the world, but hypothetical constructs whose very meaning is given within the mode of theoretical comprehension. For similar reasons, one should not expect a configurational understanding of the relation of substance and quality, since these are categories which may or may not (depending on the categoreal scheme adopted) inform the experience of concrete particulars but are not themselves objects of experience.
Each mode of comprehension tends to generate its own form of discourse, including concepts which take their proper meaning from the way in which they function within the mode. And since these concepts often have the same name as concepts peculiar to a different mode, there occurs a transformation of meaning from mode to mode and hence a systematic ambiguity of these concepts which makes misunderstanding inevitable at the same time that it conceals its existence and reason. This occurs both in the case of first-order concepts, such as "man" or "force," and of second-order concepts, such as "fact" or "theory." It is with respect to the latter that the most serious misunderstandings occur, with natural scientists regarding the "theories" of historians as unscientific, or philosophers regarding the "theories" of science as incomplete. By recognizing fundamental and irreducible modes of comprehension, it is possible to explain why such misunderstandings occur but not to eliminate the cause.
Is unity of science possible? No, because the several modes of comprehension generate and justify several methods. Is unity of knowledge possible? Yes, as knowledge of a world for a mind whose mode of comprehension gives structure to that world. The limits of the world are the limits of the discipline we adopt to inquire into it. The disciplines of inquiry are distinguished by the modes of comprehension at which they aim.