It is certain that the deliberation in such cases must be guided at once by considerations of the extent of the injury, and of the restrictions on freedom implied in the given law. But the proper estimation of these does not admit, properly speaking, of any general rule; and all calculations of probability are eminently fallacious. Theory therefore can only point out these moments of deliberation. In the reference to practice, I am of opinion that special circumstances should be chiefly regarded, and not so much the general cases; and that only when observation of the past and considerations of the present combine to represent a restriction as indispensable, should it ever be resolved on. The right of nature, when applied to the social life of a number of men, defines the boundary lines unmistakably. It condemns all actions in which, with his own fault, one man encroaches on the due province of another, and hence, includes all those cases in which the injury strictly arises from a blamable oversight, or where it is always associated with the action, or with such a degree of probability in the consequence, that the agent either perceives it or at least becomes accountable by overlooking it. In all other cases the injury proceeds from chance, and of course the agent is not bound to repair its effects. Any wider application than this, could only be gained from the tacit agreement of those living together; and this is again something positive. But that the State should rest here seems justly questionable; especially when we consider the importance of the injury to be apprehended, and the possibility of rendering the restriction imposed on freedom, only moderately hurtful to the citizens. In such a case it is clear that the right is undeniable on the part of the State, since it is to provide for security, not only in so far as the enforcement of reparation is concerned where right has really been violated, but also in adopting means for pre-
