In the federal Convention of 1787, James Madison was the acknowledged leader of the group favoring a strong central government. The final draft of the Constitution did not fully satisfy Madison, for he thought the proposed government neither strong enough in itself nor able to "prevent . . . local mischiefs," such as Shays's Rebellion. Writing from New York on October 24, 1787, he sent a copy of the Constitution along with the following letter to Thomas Jefferson, in Paris, in which he gave a summary of his impressions of the work of the Convention. In Jefferson's words the letter revealed "the rich resources of his 【Madison's】luminous and discriminating mind." Madison's statements in this letter substantially foreshadowed the arguments he was to make in favor of the Constitution in his famous tenth Federalist.

You will herewith receive the result of the Convention, which continued its session till the 17th of September. I take the liberty of making some observations on the subject, which will help to make up a letter if they should answer no other purpose.

It appeared to be the sincere and unanimous wish of the Convention to cherish and preserve the Union of the states. No proposition was made, no suggestion was thrown out in favor of a partition of the empire into two or more confederacies.

It was generally agreed that the objects of the Union could not be secured by any system founded on the principle of a confederation of sovereign states. A voluntary observance of the federal law by all the members could never be hoped for. A compulsive one could evidently never be reduced to practice; and if it could, involved equal calamities to the innocent and the guilty, the necessity of a military force, both obnoxious and dangerous, and, in general, a scene resembling much more a civil war than the administration of a regular government.

Hence was embraced the alternative of a government which, instead of operating on the states, should operate without their intervention on the individuals composing them; and hence the change in the principle and proportion of representation.

This groundwork being laid, the great objects which presented themselves were: (1) To unite a proper energy in the executive and a proper stability in the legislative departments, with the essential characters of republican government. (2) To draw a line of demarcation which would give to the general government every power requisite for general purposes, and leave to the states every power which might be most beneficially administered by them. (3) To provide for the different interests of different parts of the Union. (4) To adjust the clashing pretensions of the large and small states.

Each of these objects was pregnant with difficulties. The whole of them together formed a task more difficult than can be well-conceived by those who were not concerned in the execution of it. Adding to these considerations the natural diversity of human opinions on all new and complicated subjects, it is impossible to consider the degree of concord which ultimately prevailed as less than a miracle.

The first of these objects, as respects the executive, was peculiarly embarrassing. On the question whether it should consist of a single person or a plurality of coordinate members, on the mode of appointment, on the duration in office, on the degree of power, on the reeligibility--tedious and reiterated discussions took place. The plurality of coordinate members had finally but few advocates. Governor Randolph was at the head of them. The modes of appointment proposed were various: as by the people at large, by electors chosen by the people, by the executives of the states, by the Congress; some preferring a joint ballot of the two houses; some, a separate concurrent ballot, allowing to each a negative on the other house; some, a nomination of several candidates by one house, out of whom a choice should be made by the other. Several other modifications were started. The expedient at length adopted seemed to give pretty general satisfaction to the members. As to the duration in office, a few would have preferred a tenure during good behavior; a considerable number would have done so in case an easy and effectual removal by impeachment could be settled.

It was much agitated whether a long term, seven years for example, with a subsequent and perpetual ineligibility, or a short term, with a capacity to be reelected, should be fixed. In favor of the first opinion were urged the danger of a gradual degeneracy of reelections from time to time, into first a life and then a hereditary tenure, and the favorable effect of an incapacity to be reappointed on the independent exercise of the executive authority. On the other side, it was contended that the prospect of necessary degradation would discourage the most dignified characters from aspiring to the office; would take away the principal motive to the faithful discharge of its duties--the hope of being rewarded with a reappointment; would stimulate ambition to violent efforts for holding over the constitutional term; and instead of producing an independent administration and a firmer defense of the constitutional rights of the department would render the officer more indifferent to the importance of a place which he would soon be obliged to quit forever, and more ready to yield to the encroachments of the legislature, of which he might again be a member.

The questions concerning the degree of power turned chiefly on the appointment to offices and the control on the legislature. An absolute appointment to all offices, to some offices, to no offices formed the scale of opinions on the first point. On the second, some contended for an absolute negative as the only possible means of reducing to practice the theory of a free government, which forbids a mixture of the legislative and executive powers. Others would be content with a revisionary power to be overruled by three-fourths of both houses. It was warmly urged that the judiciary department should be associated in the revision. The idea of some was that a separate revision should be given to the two departments; that if either objected, two-thirds, if both, three-fourths, should be necessary to overrule. . . .

The second object, the due partition of power between the general and local governments was, perhaps of all, the most nice and difficult. A few contended for an entire abolition of the states, some, for indefinite power of legislation in the Congress, with a negative on the laws of the states; some, for such a power without a negative; some, for a limited power of legislation, with such a negative; the majority, finally, for a limited power without the negative. The question with regard to the negative underwent repeated discussions, and was finally rejected by a bare majority. As I formerly intimated to you my opinion in favor of this ingredient, I will take this occasion of explaining myself on the subject. Such a check on the states appears to me necessary (1) to prevent encroachments on the general authority; (2) to prevent instability and injustice in the legislation of the states.

1. Without such a check in the whole over the parts, our system involves the evil of imperia in imperio 【governments within a government】. If a complete supremacy somewhere is not necessary in every society, a controlling power at least is so, by which the general authority may be defended against encroachments of the subordinate authorities, and by which the latter may be restrained from encroachments on each other. . . .

We find the representatives of counties and corporations in the legislatures of the states much more disposed to sacrifice the aggregate interest, and even authority, to the local views of their constituents than the latter to the former. I mean not by these remarks to insinuate that an esprit de corps will not exist in the national government, or that opportunities may not occur of extending its jurisdiction in some points. I mean only that the danger of encroachments is much greater from the other side, and that the impossibility of dividing powers of legislation in such a manner as to be free from different constructions by different interests, or even from ambiguity in the judgment of the impartial, requires some such expedient as I contend for. . . .

2. A constitutional negative on the laws of the states seems equally necessary to secure individuals against encroachments on their rights. The mutability of the laws of the states is found to be a serious evil. The injustice of them has been so frequent and so flagrant as to alarm the most steadfast friends of republicanism. I am persuaded I do not err in saying that the evils issuing from these sources contributed more to that uneasiness which produced the Convention, and prepared the public mind for a general reform, than those which accrued to our national character and interest from the inadequacy of the Confederation to its immediate objects. A reform, therefore, which does not make provision for private rights must be materially defective. The restraints against paper emissions and violations of contracts are not sufficient. Supposing them to be effectual as far as they go, they are short of the mark. Injustice may be effected by such an infinitude of legislative expedients that, where the disposition exists, it can only be controlled by some provision which reaches all cases whatsoever. The partial provision made supposes the disposition which will evade it.

It may be asked how private rights will be more secure under the guardianship of the general government than under the state governments, since they are both founded on the republican principle which refers the ultimate decision to the will of the majority, and are distinguished rather by the extent within which they will operate than by any material difference in their structure. A full discussion of this question would, if I mistake not, unfold the true principles of republican government and prove, in contradiction to the concurrent opinions of the theoretical writers, that this form of government, in order to effect its purposes, must operate not within a small but an extensive sphere. I will state some of the ideas which have occurred to me on this subject.

Those who contend for a simple democracy, or a pure republic, actuated by the sense of the majority and operating within narrow limits, assume or suppose a case which is altogether fictitious. They found their reasoning on the idea that the people composing the society enjoy not only an equality of political rights but that they have all precisely the same interests and the same feelings in every respect. Were this in reality the case, their reasoning would be conclusive. The interest of the majority would be that of the minority, also; the decisions could only turn on mere opinion concerning the good of the whole, of which the major voice would be the safest criterion; and within a small sphere, this voice could be most easily collected and the public affairs most accurately managed.

We know, however, that no society ever did, or can, consist of so homogeneous a mass of citizens. In the savage state, indeed, an approach is made toward it, but in that state little or no government is necessary. In all civilized societies distinctions are various and unavoidable. A distinction of property results from that very protection which a free government gives to unequal faculties of acquiring it. There will be rich and poor; creditors and debtors; a landed interest, a monied interest, a mercantile interest, a manufacturing interest. These classes may again be subdivided according to the different productions of different situations and soils, and according to different branches of commerce and of manufactures. In addition to these natural distinctions, artificial ones will be founded on accidental differences in political, religious, or other opinions, or an attachment to the persons of leading individuals. However erroneous or ridiculous these grounds of dissension and faction may appear to the enlightened statesman or the benevolent philosopher, the bulk of mankind, who are neither statesmen nor philosophers, will continue to view them in a different light.

It remains, then, to be inquired whether a majority having any common interest, or feeling any common passion, will find sufficient motives to restrain them from oppressing the minority. An individual is never allowed to be a judge, or even a witness, in his own cause. If two individuals are under the bias of interest or enmity against a third, the rights of the latter could never be safely referred to the majority of the three. Will 2,000 individuals be less apt to oppress 1,000, or 200,000, 100,000?

Three motives only can restrain in such cases. (1) A prudent regard to private or partial good, as essentially involved in the general and permanent good of the whole. This ought, no doubt, to be sufficient of itself. Experience, however, shows that it has little effect on individuals, and perhaps still less on a collection of individuals, and least of all on a majority with the public authority in their hands. If the former are ready to forget that honesty is the best policy, the last do more; they often proceed on the converse of the maxim that whatever is politic is honest. (2) Respect for character. This motive is not found sufficient to restrain individuals from injustice, and loses its efficacy in proportion to the number which is to divide the pain or the blame. Besides, as it has reference to public opinion, which is that of the majority, the standard is fixed by those whose conduct is to be measured by it. (3) Religion. The inefficacy of this restraint on individuals is well-known. The conduct of every popular assembly, acting on oath, the strongest of religious ties, shows that individuals join without remorse in acts against which their consciences would revolt if proposed to them separately in their closets. When, indeed, religion is kindled into enthusiasm, its force, like that of other passions, is increased by the sympathy of a multitude. But enthusiasm is only a temporary state of religion, and while it lasts will hardly be seen with pleasure at the helm. Even in its coolest state, it has been much oftener a motive to oppression than a restraint from it.

If, then, there must be different interests and parties in society, and a majority, when united by a common interest or passion, cannot be restrained from oppressing the minority, what remedy can be found in a republican government where the majority must ultimately decide, but that of giving such an extent to its sphere that no common interest or passion will be likely to unite a majority of the whole number in an unjust pursuit? In a large society, the people are broken into so many interests and parties that a common sentiment is less likely to be felt, and the requisite concert less likely to be formed by a majority of the whole. The same security seems requisite for the civil as for the religious rights of individuals. If the same sect form a majority and have the power, other sects will be sure to be depressed. Divide et impera 【divide and conquer】, the reprobated axiom of tyranny, is, under certain qualifications, the only policy by which a republic can be administered on just principles.

It must be observed, however, that this doctrine can only hold within a sphere of a mean extent. As in too small a sphere oppressive combinations may be too easily formed against the weaker party, so in too extensive a one a defensive concert may be rendered too difficult against the oppression of those entrusted with the administration. The great desideratum in government is so to modify the sovereignty as that it may be sufficiently neutral between different parts of the society to control one part from invading the rights of another, and at the same time sufficiently controlled itself from setting up an interest adverse to that of the entire society. . . .

Begging pardon for this immoderate digression, I return to the third object above mentioned: the adjustments of the different interests of different parts of the continent. Some contended for an unlimited power over trade, including exports as well as imports, and over slaves as well as other imports; some, for such a power, provided the concurrence of two-thirds of both houses were required; some, for such a qualification of the power, with an exemption of exports and slaves; others, for an exemption of exports only. The result is seen in the Constitution. South Carolina and Georgia were inflexible on the point of the slaves.

The remaining object created more embarrassment and a greater alarm for the issue of the Convention than all the rest put together. The little states insisted on retaining their equality in both branches, unless a complete abolition of the state governments should take place; and made an equality in the Senate a sine qua non. The large states, on the other hand, urged that as the new government was to be drawn principally from the people immediately, and was to operate directly on them not on the states; and, consequently, as the states would lose that importance which is now proportioned to the importance of their voluntary compliance with the requisitions of Congress, it was necessary that the representation in both houses should be in proportion to their size. It ended in the compromise which you will see, but very much to the dissatisfaction of several members from the large states.