To neglect the consideration of evidence, is to neglect the essential conditions of judgment. To consider evidence duly, is to lay ourselves under a necessity of judging correctly.

§ 104. Evidence may be distinguished as real or supposed, complete or imcomplete.

Real evidence is that which sustains the judgments supposed to be deducible from it; and from which those judgments are capable of being legitimately inferred.

Supposed evidence is that which is supposed to be a legitimate basis, from which to draw certain inferences when it is not. Many of the facts attested in courts of justice, and many of the arguments adduced by controversialists, are of this description.

Evidence is complete when it establishes the judgment based upon it; or when it establishes all the parts of the judgment as truths.

Evidence is incomplete when it does not establish the judgments based on it, but only renders them more or less probable. It may be incomplete, in some one article, in several articles, or in all. For example: the evidence that a particular event will occur to-morrow, or at any other particular time, may be incomplete, in not showing that the event will occur at the time specified; or in not showing this conclusively; or it may be incomplete in not showing that the event referred to will occur at all; or that any event will occur. So evidence that a particular event did occur at a particular time, may be incomplete, in not showing that the event occurred at the time specified; or in not showing this conclusively; or in not showing that the event in question ever occurred.

§ 105. Judgments are conformable to evidence, when they are what that evidence is adapted to make them, and neither more nor less. We know that they are of this character, when the evidence in question produces them uniformly, and when it does this in conformity with our modes of judg ing in analogous cases, and are still farther confirmed in this when we discover our own judgments to bear the same relations to given evidences which the judgments of others do.

Judgments are not conformable to evidence, when they are not what evidence is adapted to make them. We know that they are not of this character, when the evidence in question does not produce them uniformly, and when it

does not produce them in conformity with our established modes of judgment in analogous cases. We are still farther confirmed in this, when we discover our own occasional judgments in question, not to bear the same relations to given evidences which the judgments of others generally do, who are equally well informed with ourselves. Correct judgments are confirmed by their agreement with previous and later judgments, depending on similar grounds; and incorrect judgments invalidated by their corresponding disagreements.

Our voluntary control over our judgments is similar to that which we exercise over our sensations, and is subject to similar restrictions. Both faculties are liable to abuse and perversion, but neither of them admits of being perverted only to a limited extent. There are sensations which we cannot but feel in circumstances appropriate to the experience of them, and judgments which we cannot but form.

§ 106. In popular and judicial language evidence is considered as capable of being measured by degrees, and persuasion, or belief which is founded on it, is considered as admitting of a corresponding measurement. One, or a few degrees of evidence, are considered as not furnishing adequate grounds of inference, when several degrees would be sufficient for this purpose. Degrees, however, are not strictly predicable either of evidence or belief.

Evidence does not properly admit of being measured or estimated by degrees. It either establishes certain conclusions or it does not establish them. Those conclusions which it establishes, it perfectly establishes. Those which it does not establish, it does not establish, and they ought not on such grounds, to be received as true, or to be adopted as judgments. To infer from given evidence, judgments which they do not require and establish, is to judge incorrectly; or to form judgments which are not conformable to evidence.

Those supposed degrees of evidence in favor of particular judgments, which do not fully establish them, are falsely called evidence. They approximate towards the nature of evidence, in the same degrees in which they are supposed to be of the nature of it; when they become legitimate grounds of judg ment, they become evidence; and all those judgments deduced from them which are not legitimate are erroneous; that is, not according to the evidence.

§ 107. Evidence may be more or less obvious, but its relations to judgments is, in all cases, the same, and does not admit of being compared in respect to degrees. Perfect conviction does not vary in degree; and imperfect conviction is not conviction in any degree. To be persuaded in any degree less than perfectly, is really to be not persuaded at all; and to believe in any degree less than confidently, is really not to believe at all.

Persons never speak of degrees of evidence or of persuasions, when they have certain convictions. Ideas which do not rest on satisfactory and decisive evidence, are the only class of judgments which we consider as capable of being measured by degrees. We have ideas that certain events will take place to-morrow. Whether they will really take place or not, we have no means of determining with certainty. We judge that they are possible, perhaps probable, or more likely, so far as as we have the means of judging, to occur than not to occur. Perhaps there is a strong probability, so far as our knowledge extends, that they will occur. We may see strong reasons to believe that they will occur, and none to believe that they will not occur. At the same time, we may know perfectly that our knowledge does not extend to all the causes, by the operation of which the occurrence of these events is to be determined. We, therefore, cannot judge that they will occur, and can only judge that their occurrence is probable.

§ 108. When we have no means of determining whether events will really occur or not, from an incapacity of investigating all the causes that concur in producing them, we may still investigate the causes within the sphere of our knowledge, and ascertain whether, so far as these are concerned, the occurrence or non-occurrence of the events in question is probable. In cases of this kind, we form certain judgments of such a kind as the evidences sustain. The evidences do not reveal the occurence or non-occurrence of the event to which they relate, considered absolutely, but its occurrence or non-occurrence, so far as particular causes are concerned, which come within the sphere of our knowledge. What the evidence proves, it proves conclusively. What it leaves unproved, we are incapable of concluding with certainty, or making an object of legitimate judgments.

§ 109. We have numerous ideas respecting objects which are not legitimate judgments. Many of these are described

as acts of persuasion and belief, when if strictly examined, they will be found not to be conformable to evidence, and to be capable of being clearly distinguished from those ideas which have substantial grounds of inference. We distinguish this class of ideas from others, as less certain, or in some degree uncertain; and consider them as more or less certain, according as the ideas from which they are deduced, approximate more or less to the character of complete evidence. In all cases in which the evidence of particular facts is in any degree incomplete, we are incapable of forming legiti mate judgments that they have occurred. They may have occurred, but their occurrence, in these cases, cannot be objects of legitimate judgment. We may judge that events have occurred without complete evidence of the facts; and these judgments may agree with the facts, but do not agree with the evidence, which by the supposition is not complete; the judgments, therefore, are not legitimate. When the evidence of facts is incomplete, we cannot form legitimate judgments respecting them. We may believe or not believe them, but have no warrant for belief or unbelief. We ought to judge them to be true, only as far as we have evidence in their favor, and to be more or less doubtful as far as any items are wanting to render the mass of evidence complete.

§ 110. A legitimate judgment that any truth is probable, of which we have no certain evidence, implies the following conditions:

1. A knowledge of an imperfect mass of evidence, which is, on the whole, favorable to the supposition in question.

2. A knowledge of the fact, that the evidence in question is imperfect, and that some other items must be added to it, to make it decisive of the fact which it is considered as rendering probable.

3. An impartial estimate of the evidence possesed, and an inference respecting the fact in question from this evidence, considered without respect to the supplement wanting to render it complete.

All judgments of this kind are liable to be erroneous in cases where they are perfectly legitimate. A given mass of imperfect evidence may be in favor of one conclusion, when the same evidence, considered in connection with the other parts necessary to render it complete, would require an op

posite one. Legitimate judgments, founded on incomplete evidence, are often reversed by later judgments, deduced from the same evidences considered in connection with others, when rendered complete.

§ 111. Where the evidence which sustains our judgments, is in any degree incomplete, we cannot form legitimate absolute judgments. We may form absolute judgments on the ground of incomplete evidence, and they may happen to be in accordance with facts, and may be confirmed and rendered legitimate and certain, whenever the evidence on which they rest is completed. But as long as the evidence on which they rest is incomplete, they are not legitimate judgments. Incomplete evidence cannot establish conclusions. The same evidence, however, that is incomplete, considered in respect to one conclusion, may be complete considered with respect to others. Thus a particular mass of evidence may be incomplete, considered with respect to the conclusion, that a particular man is guilty of any particular crime; and yet may be complete, considered with respect to the conclusion, that some man committed the crime in question; or that the particular man in question did something, or exercised some agency in the case.

Complete evidence may be redundant. So may particular parts of that which is incomplete. Redundant evidence, however, adds nothing to the legitimacy of our judgments

or conclusions.

Truths in geometry are no more true, on account of being demonstrated by evidence derived from different sources, than they would be if capable only of single demonstrations. The same is true respecting all judgments of what is, what has been, and what is to be. A past or future event, which is indicated by a single mass of complete evidence, is as certain as one which is indicated with equal clearness, by two, five, or ten such masses of evidence. There is no difference between necessary and contingent truths in this respect. Complete evidence of contingent truths, is as decisive and satisfactory in regard to them, as complete evidence of necessary truths is in regard to them.

Belief and faith are appropriate titles of valid judgments; and believing is synonymous with judging. We believe on the ground of evidence, and our belief is legitimate or illegitimate, according as it is conformable to the evidence on which it rests, or not. All that has been said of judgments