In the Introduction to A Treatise on Political Economy the author, Jean-Baptiste Say (1767-1832), makes some insightful comments on all science:
I.1 A SCIENCE only advances with certainty, when the plan of inquiry and the object of our researches have been clearly defined; otherwise a small number of truths are loosely laid hold of, without their connexion being perceived, and numerous errors, without being enabled to detect their fallacy. … I.5 The wide range taken into the field of pure politics, whilst investigating the subject of political economy, seemed to furnish a much stronger reason for including in the same inquiry agriculture, commerce and the arts, the true sources of wealth, and upon which laws have but an accidental and indirect influence. Thence what interminable digressions! If, for example, commerce constitutes a branch of political economy, all the various kinds of commerce form a part; and as a consequence, maritime commerce, navigation, geography—where shall we stop? All human knowledge is connected. Accordingly, it is necessary to ascertain the points of contact, or the articulations by which the different branches are united; by this means, a more exact knowledge will be obtained of whatever is peculiar to each, and where they run into one another. I.6 In the science of political economy, agriculture, commerce and manufactures are considered only in relation to the increase or diminution of wealth, and not in reference to their processes of execution. This science indicates the cases in which commerce is truly productive, where whatever is gained by one is lost by another, and where it is profitable to all; it also teaches us to appreciate its several processes, but simply in their results, at which it stops. Besides this knowledge, the merchant must also understand the processes of his art. He must be acquainted with the commodities in which he deals, their qualities and defects, the countries from which they are derived, their markets, the means of their transportation, the values to be given for them in exchange, and the method of keeping accounts. I.7 The same remark is applicable to the agriculturist, to the manufacturer, and to the practical man of business; to acquire a thorough knowledge of the causes and consequences of each phenomenon, the study of political economy is essentially necessary to them all; and to become expert in his particular pursuit, each one must add thereto a knowledge of its processes. These different subjects of investigation were not, however, confounded by Dr. Smith; but neither he, nor the writers who succeeded him, have guarded themselves against another source of confusion, here important to be noticed, inasmuch as the developments resulting from it, may not be altogether unuseful in the progress of knowledge in general, as well as in the prosecution of our own particular inquiry. I.8 In political economy, as in natural philosophy, and in every other study, systems have been formed before facts have been established; the place of the latter being supplied by purely gratuitous assertions. More recently, the inductive method of philosophizing, which, since the time of Bacon, has so much contributed to the advancement of every other science, has been applied to the conduct of our researches in this. The excellence of this method consists in only admitting facts carefully observed, and the consequences rigorously deduced from them; thereby effectually excluding those prejudices and authorities which, in every department of literature and science, have so often been interposed between man and truth. But, is the whole extent of the meaning of the term, facts, so often made use of, perfectly understood? I.9 It appears to me, that this word at once designates objects that exist, and events that take place; thus presenting two classes of facts: it is, for example, one fact, that such an object exists; another fact, that such an event takes place in such a manner. Objects that exist, in order to serve as the basis of certain reasoning, must be seen exactly as they are, under every point of view, with all their qualities. Otherwise, whilst supposing ourselves to be reasoning respecting the same thing, we may, under the same name, be treating of two different things. I.10 The second class of facts, namely, events that take place, consists of the phenomena exhibited, when we observe the manner in which things take place. It is, for instance, a fact, that metals, when exposed to a certain degree of heat, become fluid. I.11 The manner in which things exist and take place, constitutes what is called the nature of things; and a careful observation of the nature of things is the sole foundation of all truth.
I.12 Hence, a twofold classification of sciences; namely, those which may be styled descriptive, which arrange and accurately designate the properties of certain objects, as botany and natural history; and those which may be styled experimental, which unfold the reciprocal action of substances on each other, or in other words, the connexion between cause and effect, as chemistry and natural philosophy. Both departments are founded on facts, and constitute an equally solid and useful portion of knowledge. Political economy belongs to the latter; in showing the manner in which events take place in relation to wealth, it forms a part of experimental science.*3 I.13 But facts that take place may be considered in two points of view; either as general or constant, or as particular or variable. General facts are the results of the nature of things in all analogous cases; particular facts as truly result from the nature of things, but they are the result of several operations modified by each other in a particular case. The former are not less incontrovertible than the latter, even when apparently they contradict each other. In natural philosophy, it is a general fact, that heavy bodies fall to the earth; the water in a fountain, nevertheless, rises above it. The particular fact of the fountain is a result wherein the laws of equilibrium are combined with those of gravity, but without destroying them. I.14 In our present inquiry, the knowledge of these two classes of facts, namely, of objects that exist and of events that take place, embraces two distinct sciences, political economy and statistics. I.15 Political economy, from facts always carefully observed, makes known to us the nature of wealth; from the knowledge of its nature deduces the means of its creation, unfolds the order of its distribution, and the phenomena at tending its destruction. It is, in other words, an exposition of the general facts observed in relation to this subject. With respect to wealth, it is a knowledge of effects and of their causes. It shows what facts are constantly conjoined with; so that one is always the sequence of the other. But it does not resort for any further explanations to hypothesis: from the nature of particular events their concatenations must be perceived; the science must conduct us from one link to another, so that every intelligent understanding may clearly comprehend in what manner the chain is united. It is this which constitutes the excellence of the modern method of philosophizing.