of exercising careful judgment upon them."

In this statement, you will find, I believe, one of the essential reasons why a man of Mr. Brandeis's imaginative power has turned against the modern trust. He does not believe that men can deal efficiently with the scale upon which the modern business world is organized. He has said quite frankly, that economic size is in itself a danger to democracy. This means, I take it, that American voters are not intelligent enough or powerful enough to dominate great industrial organizations. So Mr. Brandeis, in company with many important thinkers the world over, has turned de-centralizer. The experience of history justifies his position in many respects: there is no doubt that an organization like the Holy Roman Empire was too large for the political capacity of human beings. It is probably true that the Morgan empire had become unwieldy. It may be that the Steel Trust is too large for efficiency. The splendid civilizations of the past have appeared in small cities. To-day

if you go about the world you find that the small countries like Belgium, Holland, Denmark, are the ones that have come nearest to a high level of social prosperity. I once heard George Russell (E), the Irish poet and reformer, say that an ideal state would be about the size of County Cork.

Yet it is not very helpful to insist that size is a danger, unless you can specify what size.

The senators asked Mr. Brandeis that question. They pressed him to state approximately what percentage of an industry he considered an effective unit. He hesitated between ten per cent. and forty per cent., and could not commit himself. Obviously, for how could Mr. Brandeis be expected to know? Adam Smith thought the corporations of his day doomed to failure on the very same grounds that Mr. Brandeis urges against the modern corporation. Now the million dollar organization is not too large for efficiency and the billion dollar one may be. The ideal unit may fall somewhere

between? Where? That is a problem which experiments alone can decide, experiments conducted by experts in the new science of administration.

The development of that science is the only answer to the point Mr. Brandeis raises. Remarkable results have already been produced. Every one of us, for example, must wonder at times how the President of the United States ever does all the things the papers say he does. When, for example, does the man sleep? And is he omniscient? The fact is that administration is becoming an applied science, capable of devising executive methods for dealing with tremendous units. No doubt the President with his increasing responsibilities is an overworked man. No doubt there are trusts badly administered. No doubt there are inflated monopolies created for purely financial reasons. But just what the limits of administrative science are, a legislature is no more capable of determining than was Mr. Brandeis. Only experience, only trial and in

genuity, can demonstrate, and in a research so young and so swift in its progress, any effort to assign by law an arbitrary limit is surely the most obvious meddling. Say to-day that one unit of business is impossible, to-morrow you may be confronted with an undreamt success. Here if anywhere is a place where negative prophecy is futile. It is well to remember the classic case of that great scientist Simon Newcomb, who said that man would never fly. Two years before that statement was made, the Wright brothers had made secret flights.

It may well be that the best unit is smaller than some of the modern trusts. It does not follow that we must break up industry into units of administration whose ideal efficiency is spent in competing with one another. I can understand, for example, the desire of many people to see Europe composed of a larger number of small nations. But I take it that everyone wishes these small nations to coöperate in the creation of a common European civilization. So it

may

is with business. The unit of administration be whatever efficiency demands. It may be that the steel industry would gain if it were conducted by forty corporations. But at the same time there are advantages in common action which we cannot afford to abandon. Technical improvement must be for the whole industry, the labor market must be organized and made stable, output must be adjusted to a common plan. The appearance of federal organization seems to suggest a possible compromise in which the administrative need for decentralization is combined with the social demand for a unified industrial policy.

No one, surely, proposes to revive the little business monarch who brooded watchfully over every operation in factory and office, called his workingmen by their pet names, and was impelled at almost every turn by Adam Smith's "natural propensity to truck and barter." For just as in political government "the President" does a hundred things every day he may never even