singly or in bodies, but abstractions, it is impossible to try any proposition by a historical standard, until it has been translated out of this symbolical language into a dramatic form, in which real men bear a part.

This view of the subject is adopted by M. Guizot, in his lectures on the history of modern civilization, and, indeed, governs their arrangement and contents. He begins by laying it down, that there is a certain civilization common to the states of Western Europe, and that it consists of scattered elements, which must be sought in the history of the different countries. He devotes fourteen lectures, forming one volume of his work, to the leading causes of European civilization:(264) but, having completed this general survey, from the fall of the Roman empire to the French revolution, he declares that, as soon as the historian descends from the most general facts, the unity of his subject disappears; and that in order to trace the progress of civilization in detail, it is necessary for him to narrow the field of inquiry.

militaire, j'ai jugé convenable, pour plus de clarté, de me rapprocher davantage des formes de l'appréciation concrète. Mais il importe à notre but principal de reconnoître directement que je ne me suis ainsi nullement écarté, au fond, du caractère abstrait indispensable à une telle opération, suivant les explications préliminaires du chapitre précédent. Car, ces dénominations de Grec et Romain ne désignent point ici essentiellement des sociétés accidentelles et particulières; elles se rapportent surtout à des situations nécessaires et générales, qu'on ne pourrait qualifier abstraitement que par des locutions trop compliquées' (ib. p. 265). According to this explanation, the words Greek and Roman do not signify the nations so called they are mere algebraic symbols, to denote certain abstract communities. See also p. 268, note, and p. 295.

(264) Concerning the existence of a common European civilization, see Guizot, Hist. Gen. de la Civilisation en Europe, leçon i. In the first lecture of his course, on the 'Histoire de la Civilisation en France,' he thus expresses himself:- Nous avons reconnu, l'été dernier, qu'il y avait une véritable unité dans la civilisation Européenne; mais cette unité n'éclate que dans les faits généraux, dans les grands résultats. Il faut s'élever au haut des montagnes pour faire disparaître les inégalités, les diversités du territoire, et découvrir l'aspect général, la physionomie essentielle et simple, de tout le pays. Quand on sort des faits généraux, quand on veut pénétrer dans les particularités, l'unité s'efface, les diversités se retrouvent, on se perd dans la variété des évènements, des causes, des effets; en sorte que, pour raconter l'histoire avec détail, et y conserver, cependant, quelque ensemble, il faut absolument en rétrécir le champ.' Compare his seventh lecture on European civilization, where he remarks, that his general sketch of the origin of the communes, or free towns, does not apply indifferently to the free towns of each country. Certain features are common to all, but the differences are wide and important.

For this reason he devotes the rest of his course, consisting of forty-nine lectures, distributed over five volumes, to the history of civilization in France exclusively.

§ 25 It appears, from the preceding remarks, that the essence of political history is to be a register or record, including nothing false, and omitting nothing important with reference to its end. It is a collection of observed facts, arranged in a proper chronological series, so as to show their mutual sequence or dependence, and subjected to a proper process of selection, for the sake of rejecting everything trifling or irrelevant. In so far as a history is a collection of observed facts, true and relative to some common subject, it is analogous to a set of observations in astronomy, or any other physical science. There is, nevertheless, a material difference between the modes of exhibiting the results of observations in politics and in physics, to which it is necessary for us to advert.

In the physical sciences, observations are sometimes made without any theoretical view, and laid before the world in their naked state, as is often the case with subjects of natural history. Sometimes, however, they are connected with their scientific employment. The observer uses them himself as a ladder, by which he may climb to a scientific conclusion. He both collects the materials and builds the edifice. But the political historian is merely a collector and arranger of materials. He constructs no political theories himself out of the materials which his industry has accumulated, which his judgment has arranged, and which his sagacity has illustrated. This constant difference between the modes of dealing with observations in physics and in politics appears to rest on two grounds, which we will attempt to explain.

The first of these grounds consists in the great difficulty and complexity of the process by which political history is formed. That which in physics constitutes, in general, only one stage, in politics requires two. In physics, when the phenomena to be observed have been determined, the main difficulty consists in providing fit instruments or methods for observing them. When

this obstacle has been effectually overcome, the mere work of observing and registering the observations (though at first much intelligence may be needed) demands little more than steady attention and mechanical regularity. The observations, as they come from the hands of the observer, are ready for the use of the theorist. Astronomical and meteorological observations, consisting of numerical data, sometimes, indeed, require a further process of digestion; but even this, when the principle is once settled, consists merely of arithmetical operations. In general, therefore, it may be said that there is no intermediate workman between the original observer and the theorist in physics. Whenever, indeed, the experimental method is applied, such a division would be impossible; for the experimenter and theorist are necessarily the same person. In all experiments, as was pointed out in the last chapter, (265) the process is intelligent, and the general conclusion is drawn by the experimenter himself.

But in politics it is otherwise. The subject-matter of observation is so much more complex than in physics, that not only is it necessary to make a constant separation between the observer and the theorist, but in general, it is further necessary to interpose an intermediate person between these two parties; and to assign a distinct and a high office to the digester of the observed facts. This intermediate person is properly the historian. He stands between the primary observer on one side, and the theorist on the other. As the facts narrated by the historian occur over a wide surface, they cannot be observed by any one person. To supply the facts described in a volume of national history, the observations of large numbers of persons must contribute. Even to form the contents of a single newspaper, how many persons-some employed, and some communicating voluntarily-in various parts of the country, each subscribe their quota. In order to lay the foundation of a historical narrative, it is necessary to obtain the services of short-hand writers, and other reporters of public proceedings; keepers of

(265) Above, ch. vi. § 2.

public records and parliamentary journals; registrars of births, deaths, and marriages; heralds and genealogists; authors of statistical accounts of population, revenue, and expenditure, trade, and the like; collectors of state-papers and public documents; and other compilers of the information which constitutes the skeleton and framework of history. Next to these come the writers of biographies, memoirs, and other narratives of a historical character, but too confined in their subject to deserve the name of history.(26) Such, for instance, are the lives of great men by Plutarch and Cornelius Nepos;(7) such the numerous French memoirs, and journals of military and naval operations, as a siege, a battle, or a campaign. Besides subsidiary works, avowedly intended as contributions to history, information convertible to the use of the historian may be gleaned from a vast variety of writings, though not designed for his service. Speeches, pamphlets, reviews, magazines, poems, plays, ballads, letters, and numerous other publications, contain incidental allusions to the events of the day, and therefore may furnish evidence upon matters of fact. The former documents which we enumerated are, for the most part, intended to serve as historical records. They are registered with a view of perpetuating the memory of the events. But official reports, coins, allusions in contemporary satires, and the like, though they furnish historical information, are not intended to serve as historical records. In this respect, they resemble the fossil remains in a stratum, which serve incidentally to determine its geological age. (26)

All these are materials, some of a more solid, others of a slighter consistency, out of which the historian can construct his edifice. In proportion as the professed historical accounts are

(266) On the difference between memoirs and history, see Ranke, Zur Kritik Neuerer Geschichtschreiber, p. 163.

(267) With respect to the difference between biography and history, and as to the former being more intended for the representation of moral character, see Plutarch, Alexander, c. i.

(268) On the distinction between intended and unintended historical monuments, see Rühs, Propädeutik, p. 184. It is illustrated by the dif ference between a medal and a coin.

meagre, he is driven to these secondary resources. This is especially the case with the modern writer of Greek or Roman history, who must often rely upon incidental notices, without straining them too far, or misconstruing their effect. Now, for making a proper use of these materials, so much discrimination is requisite; so much choice in selecting and rejecting facts ; such freedom from prejudice and interest; so much judgment in appreciating men's actions, in estimating their professions, and conjecturing their motives; so much skill in co-ordinating and arranging facts; in distinguishing between causation and mere sequence or proximity, between juxtaposition and mutual dependence; and in determining the value of evidence, that it is necessary to appropriate to the task the exclusive and undivided attention of one man.

Not only accuracy and memory, but a power of seizing the material and rejecting the immaterial facts, of winnowing away the chaff, and retaining the corn, is essential to the construction of a narrative embracing a complex series of events. Before such a narrative can be constructed, the historian must frame in his own mind a certain theory or hypothesis, by which he explains the concatenation of the events. By applying this theory, he determines which facts are material, and which immaterial. Even in a narrow field of history, the res gestæ are so numerous that the sequence of the material events becomes unintelligible if no discrimination is exercised, and if everything, however unimportant and incidental, is poured out, without any restraint upon the garrulity of the narrator. Facts must be passed through a succession of sieves of increasing fineness before they are ready for history. It is this process which every litigated question undergoes before it is submitted to a court; and it is mainly for conducting this process effectually and expeditiously, that the assistance of professional advisers is called in, who are able to extract from the mass of statements those matters which are pertinent to the question at issue, and lay the matter before the court in a condensed shape, ready for decision; whereas the parties themselves would, if left to their unaided