Napoleon and some of the other despotic governments of the Continent. This system may be carried so far as to destroy all freedom of speech, and confidence of intercourse. (3) The opening of letters in the post-office, and the seizure of private papers, are likewise powers which governments in general exercise with reserve and caution, and principally in order to assist the enforcement of the criminal law.

Generally, it may be said that one of the first duties of persons invested with any political functions, is to observe accurately and extensively the subjects committed to their care. If it be the head of a government, a minister of state, or the governor of a province, he must keep an attentive eye upon all the principal facts which concern the welfare of the entire political community over which his care extends. (35) If he be charged with supreme military or naval command, the necessity of a constant, vigilant, and active superintendence is equally obvious. If he be a functionary having a more limited sphere of action, it is his duty to watch carefully all the circumstances which fall within the circle of his duties. For example, an officer of the excise must take note of all the proceedings of the manufacturer, an officer of police must take note of all the infractions of public order within his inspection. In proportion, indeed, as the scale of duties descends, the necessity for immediate and personal observation on the part of the functionary increases. Superior officers are more assisted by vicarious observation. They receive reports and accounts from various subordinate officers and departments; and are thus able to form a collective opinion upon the state of things from the testimony of numerous original witnesses. But although a person exercising high political functions does not in general derive much information from the observation of his own senses, it is, on that account, the more necessary for him to be regularly supplied with the results of the observations of others. It has long ago been

(34) See Tacit. Agric. c. 2.

(35) Ad consilium de republicâ dandum, caput est, nosse rempublicam.' -Cic. de Orat. ii. 82.

perceived that the purposes of government require not only that information should be exact, but that it should be punctually transmitted, and regularly conveyed, from the reporting functionaries to the seat of empire, or to the proper superior officer. The first institution of government posts is due to this want of a central government; posts were established both in the Persian kingdom and in the Roman empire, for the purpose of conveying intelligence from the governors of provinces to the sovereign.(3) In modern times, the mechanical improvements by which communication, both by sea and land, has been so greatly facilitated, have rendered the transmission of intelligence to the seat of government rapid, frequent, and regular, to an extent which the most active imagination in antiquity could not have contemplated. (3) Railways, steam navigation, and telegraphs, have

(36) The king's messengers, in the Persian empire, who carried his orders to the satraps in the subject provinces, were called by the Greeks ypaμμaтopópoi, or despatch-bearers.-See Plut. Cimon, 19; and Lucian, Rhet. Preceptor, c. 5, on the ypaμμaтоþópoɩ of Alexander.

(37) The slow progress of news, even on the most important political events, in the ancient states, appears from many facts mentioned incidentally by the historians. Thus Athenagoras, the popular orator at Syracuse, throws doubts on the Athenian expedition, and represents the report of it as a trick of the oligarchical party, in a speech which was probably made about the time when the armament was sailing from the Piræus, and consequently after numerous debates in the Athenian ecclesia, and a long series of preparations for war. (Thuc. vi. 36; and as to the time of the speech, see Grote, vol. vii. p. 249.) As to the account of the catastrophe at Syracuse being brought to Athens by a casual visitor, see the story in Plutarch, Nic. 30; and compare the expressions in Thuc. viii. 1. Plutarch speaks of an obscure rumour as to the capture of Rome by the Gauls having found its way to Greece in the time of Heraclides Ponticus, and of its being accurately known to Aristotle. (Camill. 22.) Now Rome was taken by the Gauls in 390 B.C., and Aristotle was only born in 384 B.C., six years after the event; and Heraclides could not have preceded him. (See Gell. N. A. xvii. 21, § 25.) Livy thinks that Alexander the Great was not known, even by report, to the Romans of his own time (ix. 18).

Cæsar, in whose time the Romans had organized a regular system of communication by means of military roads and messengers, censures the reliance of the Gauls upon unofficial rumours in matters of government :'Est autem hoc Gallica consuetudinis; uti et viatores etiam invitos consistere cogant, et quod quisque eorum de quâque re audierit aut cognoverit, quærant; et mercatores in oppidis vulgus circumsistat, quibusque ex regionibus veniant, quasque ibi res cognoverint, pronun ciarecogant. His rumoribus atque auditionibus permoti de summis sæpe rebus consilia ineunt: quorum eos e vestigio poenitere necesse est; quum incertis rumoribus serviant, et plerique ad voluntatem eorum ficta respondeant.'-De B. G. iv. 5.

reduced the obstacles to communication arising from space to a very low quantity; and in addition to the official information systematically procured by a government, much intelligence likewise reaches it by the newspapers, and other voluntary channels.

A statesman who guides the practical policy of a country in its internal and foreign relations, to whom the public look for advice and decision at seasons of distress, difficulty, doubt, and danger; when invasion is threatened, when civil dissensions arise, when food is scarce, when commercial panic rages, and the national credit is impaired is often likened to the steersman who stands at the helm, and regulates the course of the ship. Like the steersman, he can only guide the national policy by an attentive and vigilant observation of all the material facts, upon which the safety of the state depends.

The observations which are converted to the use of the historian, and those which are made for the purposes of government, bear a close resemblance to one another; inasmuch as, in both cases, the full fact is noted, with all its circumstances of agent, time, and place. Hence it is that official reports, prepared for the information of government, or evidence given in a court of justice, though intended merely for a definite practical purpose, may be afterwards employed by the historian as materials for his composition. Thus the Venetian Relazioni, and other diplomatic correspondence, and reports of state trials, become historical documents, though they had originally no historical purpose. History, however, is a narrative of the successive acts of a nation; it implies a series of events, connected either in the way of causation, or of chronological sequence; whereas the investigations of a government for practical purposes are generally much less extensive in their reach, and relate to a single transaction, or to a few consecutive and connected events.

1

§ 10 The observation, registration, and arrangement of those facts in politics which admit of being reduced to a numerical expression has been, of late years, made the subject of a distinct science, and comprehended under the designation of statis

tics. (3) Although the accounts kept for the use of a government may sometimes be called statistical, yet it is of the essence of statistics that its object is scientific, not practical; that it is intended to represent the truth of facts, not to subserve some immediate purpose of administration or legislation.(39) The cultivation of statistics has been owing, in great measure, to the increased attention which the science of political economy has received since the writings of the French economists and of Adam Smith; inasmuch as many of the most material facts relating to trade, currency, taxation, production, population, &c., admit of being represented in numbers. Nevertheless, statistics

may include all branches of political observation; for example, there are statistical accounts of crimes and punishments, of diseases and mortality, of education, and of other subjects.(*)

The facts observed and recorded by the statistician are important not only to the historian and the political philosopher, but also to the practical statesman. To the historian, statistical facts concerning the population, commerce, military and naval strength, revenue and expenditure of a country, are important, as serving to illustrate the condition of the people whose acts and destinies he narrates.(") The political speculator has frequent reference to them in establishing principles of legislation, especially in the province of political economy, where the facts are peculiarly susceptible of arithmetical notation; and although statistical accounts have properly no immediate relation to

(38) As to the origin of this term, see above, ch. iii. § 9.

(39) Upon the nature and province of the science of statistics, see the introduction to the Journal of the London Statistical Society, vol. i. 1839. The science of statistics (it is there remarked) 'does not discuss causes, nor reason upon probable effects; it seeks only to collect, arrange, and compare, that class of facts which alone [?] can form the basis of correct conclusions with respect to social and political government. . . . Its peculiarity is, that it proceeds wholly by the accumulation and comparison of facts, and does not admit of any kind of speculation.' In this respect it resembles history. It is further remarked, that the statist commonly prefers to employ figures and tabular exhibitions.'

(40) As to the increased observation and registration of economical facts by governments, see Blanqui, Histoire de l'Economie Politique, tom. ii. p. 326-9.

(41) See, for example, Mr. Macaulay's Hist. of Engl. vol. i. c. 3.

practice, yet as they correctly represent, upon a large scale, the facts with which he has to deal, the practical statesman can scarcely take a single safe step without consulting them. Whether he be framing a plan of finance, or considering the operation of an existing tax, or following the variations of trade, or tracing the influences of a poor-law, or studying the public health, or examining the effects of the criminal law, his conclusions ought to be principally guided by statistical data. It is by comparing the numbers of the subject under consideration, by noting whether they have diminished, or increased, or been stationary, and by inquiring into the causes of such a state of things, that his practical judgment ought to be formed. At times when information of this kind was not considered necessary as the foundation of political measures, legislation must often have proceeded upon casual and partial facts, upon an imperfect knowledge of the general state of things, and, in some cases, almost at hap-hazard.

The systematic observation and registration, on a large scale, of political facts which admit of being enumerated is, as we have already stated, a modern practice, even among civilized nations. Among Oriental nations it is nearly, among savage tribes it is wholly, unknown. Enumeration is a slow and laborious process; and until experience has taught us its necessity where correctness is required, there is a disposition, particularly among uncultivated people, to rely upon conjecture. To count (says Dr. Johnson) is a modern practice; the ancient method was to guess; and when numbers are guessed, they are always magnified.'(*) The first impression made upon the mind by a large number almost always exceeds the truth; so that there is a natural tendency to exaggeration, when the belief is uncorrected by enumeration. Besides, in many of the accounts of armies, and of men killed in battles, of the size of buildings, and the like, national vanity, and the love of the marvellous, contribute to swell the amount. (3) It is not until civilization has made

(42) Journey to the Hebrides; Works, vol. ix. p. 95. Gibbon somewhere calls arithmetic the natural enemy of rhetoric.

(43) See Winer, B. R. Wörterbuch in Zahlen; Niebuhr, Hist. of Rome, vol. ii. p. 69.