own devices, act accordingly, and fatten to their fill upon the commonwealth.

The inference is obvious. Men in power must not be trusted; on the contrary they must be checked, controlled, and called to account. Public functionaries, as Bentham phrases it, must be made uneasy. Confidence must be minimised, control must be maximised, and all the devices so well known to democracy, short parliaments, democratic suffrage, ballot, and so on, must be brought to bear in order to secure to the last jot and tittle responsibility to those whose interest is identical with the public good, to wit, the whole people or, at any rate, the numerical majority thereof.

It is obvious that, if this be true, Whig "trusteeship" must vanish; mankind, let alone Whig mankind, is not sufficiently disinterested. And in lieu of it comes the pure Benthamite gospel of unlimited responsibility to the people, as the one thing that can save the commonwealth from the rapacity of rulers, be they despotic, or be they democratic.

This radical doctrine, one must add, was, in Mill's hands, punctuated by the tone of its exponent. All the geniality of Bentham is gone. It was a thing unknown either in the public or the private life of this mordant disciple. Instead, we have invective and contempt for the "drivellers" who are fool enough to set their trust in Whig or Tory oligarchs. "Confidence itself is another name for scope to misrule. The author of Hudibras said well: All that the knave stands in need of is to be trusted, after that the business does itself.""

It was with the doctrine thus briefly outlined that Macaulay found himself confronted. And from the unchastened rhetoric of the Essays on Utilitarianism, in which he dealt with it, two substantial issues emerge.

Of these the one evokes Macaulay's wonted appeal to history. History, as he read it, taught that the nations of the earth had wonderfully prospered without universal suffrage and other radical panaceas. And in England, had not the Whig aristocracy done great things for the national life and liberties? Had not person and property become secure, and ever more secure? Had not despotism been checked and corruption repressed? Had not the arts and sciences gloriously flourished ? Had not industry, commerce, and comfort enormously increased? On this ground Macaulay takes his stand and refuses to move from it at the bidding of a theory "from which," as he remarks, "but for two or three passing allusions, it would not appear that the author was aware that any government actually existed among men."

This is the line of criticism that leads on Macaulay to put his finger on the weak point of Mill's theory of government. Mill's theory was rigorously deductive, and the characteristic weakness of all deductive theory-especially of the geometrical type which founds upon few and simple principles-is well known. A mistake at the outset is fatal, nor can any logical concatenation of inferences atone for fragility in the first links of the chain. Macaulay saw this clearly, and presses his advantage to the uttermost. He denies outright the possibility of deducing any theory of government from psychology. Psychology must needs abstract from all the peculiar local and even national characteristics of human nature,-human nature which, from age to age and from place to place, inevitably takes on the changing colours of changeful circumstance. It abstracts, that is to say, precisely from these conditions which must needs be of vital moment to the theoriser in government. Whereas, the kind of theory of government that really concerns the thinker in politics is not a theory

of government for man in general, but for Englishmen, or Frenchmen, or Russians, or Turks; in short, for concrete human beings, and not for the thin dehumanised shadows of a universalising psychology. This is Macaulay's drift, and, in the heat of combat, he ventures to push his argu-. ment even further. Mill, he argues, is radically wrong in making psychology his starting point at all. The more fruitful method is to begin with the larger entity; not with the individual man, but with society. First let us study the nature of the social environment, including the government under which individuals live, in order that we may thereby come to know the nature of the individuals who develop under its influence. To take the opposite course-as Mill does-to start with propositions about the individual, and then to pass on to deductions about government is, therefore, nothing less than the fallacy of making what is last first, and what is first last. "Our knowledge of human nature," so runs a quite unequivocal statement, "instead of being prior to our knowledge of the science of government, will be posterior to it."

The course of subsequent thought was destined to prove how formidable Macaulay's criticisms were. Benthamism was to produce a disciple whose salient characteristic was the ability to learn from other minds, and we have it on the evidence of this disciple himself-is it not written. in John Stuart Mill's Autobiography?—that Macaulay's criticisms completely opened his eyes, already opened partially, to the impossibility of deducing a theory of government, more geometrico, from a handful of psychological premises. Nay, they did more. When we read in the Logic of the younger Mill that there is no such thing as a "science of government" at all, because government is a fact so deeply immersed in complex and changing circumstances, we have convincing evidence that Macaulay

had virtually carried one of the central Benthamite positions. At very least he drove the younger Mill to the abandonment of that abstract deductive method that had so completely, and delusively, satisfied his father.

Yet the critic overleaps himself. As happens sometimes with enthusiasts for the historical method, he falls into the fallacy of underrating analysis. Having disposed of the "geometrical method" of Bentham and James Mill, of whom he allows himself to speak as if they were halfsimpletons and half-imposters, he is all too eager to supplant it by a pure Baconian empiricism, which would render any real explanation of social facts a permanent impossibility. For when the investigator after Macaulay's own heart had, by strict inductive process, generalised his principles, he would still be far enough from that knowledge of causes without which the explanation of social events is still to seek. Let us suppose him to have convinced himself that commercial countries are democratic, and that free-trade countries grow rich, and that light sentences diminish crime. They may all be valuable generalisations so far as they go. No one need make light of them. They may summarise large masses of facts. But when it comes to the question: Is commerce the cause of democracy? Is freedom of trade the cause of wealth? Are light sentences the cause of decrease of crime?, not only does there emerge the possibility that, in each of these cases, the effect may not be due to the supposed cause, it is even possible that the effect may have come about in despite of the supposed cause. Democracy, wealth, decrease of crime are all effects which occur under conditions of exceeding number and complexity. Some of these conditions may tend to promote, others may tend to retard them, others still may exercise but a slight influence in either way. Nor is any man, be his historical outlook

never so wide, beyond the mere beginnings of explanation till he has analysed up the concrete fact of democracy, or wealth, or crime, into its conditions, and come thereby to understand the conditions which are actually operating within the complex body politic. So far as mere induction. can pronounce, democracy might be due to many other conditions besides commerce, just as wealth might be due to the improvement of the arts, or the diminution of crime to a religious revival, or to industrial prosperity. Without analysis, that is, without knowledge of the conditions actually present in the facts before him, the observer is still groping in the dark. He does not know what it is that is really happening. Nor can he ever know till he has reinforced his observations and inductions by a social analysis as thorough-going as he can make it. Till then, he is not much better in his diagnosis of the body politic, than is the medical empiric, who is so satisfied with having established a conjunction between a symptom and a malady, that he forgets it was his first duty to come to his work with a previous knowledge of the manifold organs and functions of the body physiological.

It follows that James Mill was, after all, not so absolute a simpleton as Macaulay would have us think. He was doubtless wrong in ignoring history; he was wrong in the naive supposition that a few psychological dogmas could be the basis of a theory of government; but he was not wrong in insisting on deduction in political science. Without deduction, no such thing as real explanation is possible in social subject matter. What was needed was not a repudiation of deduction, but rather its expansion-its expansion in such a fashion that the empirical generalisations about concrete facts might be more fully considered, and, finally, deduced from the laws of all the manifold tendencies, economic, political, moral,