Brentano on Truth and Evidence;
Understanding Truth from a First-Person Point of View
In: Brentano Studien, 16, 187-205
Maria van der Schaar Leiden University
Abstract
Does the criticism raised against the use of judgemental evidence in logic and epistemology still stand, or is the notion crucial in our understanding of truth? First, Brentano’s account of truth and evidence is expounded. Then, the different kinds of criticism that may be raised against Brentano’s account of truth and evidence are presented. Finally, it is argued that, although there is a serious problem with this account of truth and evidence, a modification of the theory can be of importance for philosophy today. In Brentano’s account of truth and evidence, a first-person perspective plays a crucial role, and it is precisely this aspect we need in epistemology and in logic today.
Introduction1
The notion of evidence that forms the focus of this paper is not to be understood as a piece of evidence, as evidence for a judgement, but as evidence of a judgement, also called
judgemental evidence. The idea that judgemental evidence may play a role in logic and
epistemology has been criticised at the beginning of the twentieth century. Generally, this criticism has convinced philosophers that we may no longer use this notion of judgemental evidence. The criticism of the Neo-Kantian Leonard Nelson is especially relevant to our paper, as Nelson directs it to the writings of the Brentano School. It is found in Nelson’s Über das sogenannte Erkenntnisproblem, published in 1908, and is part of a general thesis that it is
impossible to find a criterion to distinguish knowledge from other judgements made.
Are we nevertheless entitled to use the notion of judgemental evidence
notwithstanding the criticism that has been formulated? I first (§1) expound Brentano’s account of truth and evidence. Then (§2) I present the different kinds of criticism that may be raised against Brentano’s account of truth and evidence. Finally (§3), I argue that, although there is a serious problem with this account of truth and evidence, a modification of the theory can be of importance for philosophy today. As I read Brentano’s account, a first-person perspective plays a crucial role, and it is precisely this aspect we need both in epistemology and in logic.2
1. Brentano on Truth and Evidence
To what does truth pertain? For Brentano, truth pertains to judgement, and what it means for a judgement to be true is explained in terms of its evidence. Evidence pertains only to
judgements, not to presentations or emotions (Brentano 1930, 144).3 As evidence is an epistemic notion, Brentano’s account of truth is accordingly epistemic. This holds for
2 For those interested in the historical context of Brentano’s account of truth and evidence, and in the question how Brentano’s epistemic account of truth relates to his account of the law of excluded middle, see (Schaar 2002/2003).
3 Emotions, such as loving and hating, may, though, be experienced as correct, ‘das Analogon … von der
Brentano’s later writings, as well as for some earlier phases in his thinking.4 How should we understand Brentano’s concept of truth?
For Brentano, understanding himself to be part of the Aristotelian tradition, all concepts have an empirical source; they have their source either in outer or in inner
perception. For Brentano, crucial to our understanding of a term is knowing its source. If we want to explain a concept, we need to know from which it derives. This means that in explaining a concept, we need to point to phenomena of inner or outer perception: ‘Die Aufgabe der Bestimmung eines Begriffes ist aufs engste verknüpft mit der Frage, woher er gewonnen ist. Die Erklärung eines Terminus ist in letzter Instanz der Hinweis auf gewisse Phänomene.’ (Brentano 1952, 135). ‘Terminus’ or term is here understood in its traditional sense as a general name together with the concept it expresses. It is here useful to distinguish between concepts that may be defined in terms of other, more primitive concepts, and simple concepts that cannot thus be defined (cf. Brentano 1915c, §7). Although Brentano’s
terminology is not fixed, one may distinguish between the explanation of a complex concept and the elucidation of a simple concept. For Brentano, both kinds of concepts find their origin in experience. The only way to understand a simple concept A is to elucidate it by pointing to one’s intuition of a particular case of A (die Anschauung des einzelnen, Brentano 1889a, §60).
The complex concept is understood by means of a definition or explanation in terms of primitive concepts, and in order to fully understand it, one needs to have or have had the corresponding intuitions of particular cases of these primitive concepts. Both the explanation of a complex concept and the elucidation of a simple concept are thus essentially related to intuitions in which the concept finds its origin. In order to understand the concept red, we need to have or have had the appropriate intuition; a colour-blind person cannot fully understand what red is. As Brentano puts it in the published lecture Über den Begriff der
4 Arkadiusz Chrudzimski has shown that Brentano defends an epistemic notion of truth already in his
Wahrheit (1889), after having criticised the correspondence definition of truth: if we need to
give a definition of a concept, we cannot use merely general determinations. And, especially in the case of a primitive concept, such as red, we cannot give an elucidation of it without pointing to the intuition of a particular case of red: ‘Was soll alle Verdeutlichung des
Begriffes Rot oder Blau, wenn ich jemand nicht ein Rot und ein Blau vor die Augen bringe?’
(Brentano 1889a, §60). As Brentano later puts it, if someone wants to understand the meaning of ‘red’ he has to be put in a certain position (Lage, Brentano 1966, 207), in order to be able to apprehend intuitively what the term is about. Equally, a definition of truth will not help one to understand the notion of truth, if the relevant intuition fails (idem).
The traditional correspondence theory of truth is not able to point to such an intuition.
And there are other problems with the correspondence theory, according to Brentano. If one defines truth as a correspondence between judgement and a thing or res, there is the problem (a) that correspondence cannot be taken in a literal sense as identity or similarity; (b) that there is often no actual thing (reales Ding) to which our judgement may correspond – for example, to what would a true but negative judgement correspond? And (c) if there is such a thing, the correspondence theory is not able to give us a criterion for truth. In order to
establish the correspondence between one’s judgement and the actual thing, the actual thing needs to be given to the judging agent; this means that the actual thing must already be known to the agent, which leads to an infinite regress (idem, §58, p. 28), or a circle in the explanation (cf. Kant’s Diallele in the Jäsche Logik, A 70). This is not to say that Brentano rejects the correspondence definition, it is rather that one has to clarify its scope. Here, in his lecture on truth, Brentano uses a weaker formulation of the definition: a judgement is true, if it is in accordance with the object. The definition thus says no more than ‘ein Urteil sei wahr, wenn es einen Gegenstand zutreffend beurteile, …, wenn er ist, sage, daβ er sei, oder wenn er nicht ist, sage, daβ er nicht sei.’ Brentano 1889a, §57). In Brentano’s account of judgement this
means: the judgement A exists is true precisely if A exists; the judgement A does not exist is true if A does not exist, if there are no A. This definition merely allows one to exchange the term ‘truth of an affirmative judgement J’ for the correlated term ‘existence of an A’, and to exchange the term ‘truth of a negative judgement J’ for ‘non-existence of an A’ (cf. idem,
§57). It is for this reason that Brentano calls the correspondence definition of truth a nominal definition (idem, §60). 5 As Brentano ends Über den Begriff der Wahrheit, ‘Auch jetzt, nach Ausschluβ der Miβverständnisse, würde die Definition demjenigen nichts sagen, dem die
Anschauung fehlte.’ (idem). Brentano is thus in need of a new understanding of truth, one that relates the notion to the relevant intuitions from which the concept derives.
One should not read Brentano as though he endorses the correspondence definition of truth in 1889, and then exchanges this definition for an epistemic definition of truth in his later writings. As late as 1915, Brentano writes: the agent who judges that something is, is not, is possible, etc., judges with truth, if the relevant thing is, is not, is possible, etc.
(Brentano 1915a, 139). One finds this passage just before the famous epistemic definition of truth, presented below. On Brentano’s account, the traditional correspondence formulation of truth, adaequatio rei et intellectus, has its value, if one uses it in the nominal sense explained above, although it does not give us a full understanding of the notion of truth.
To what intuition can we point in order to explain the concept of truth? In the
published lecture Vom Ursprung sittlicher Erkenntnis from the same year as Über den Begriff der Wahrheit, Brentano says: ‘Wir nennen etwas wahr, wenn die darauf bezügliche
Anerkennung richtig ist.’ (Brentano 1889b, § 23). In this lecture, Brentano starts his explanation of the notion of the good by pointing to the origin of our mental acts. We call something good, if the act of loving, the act of approval, that is related to it is right. And, Brentano adds, the notion of the good finds its origin in an act of rightly loving. So, the
5 Charles Parsons rightly says that Brentano’s reading of the definition is deflationary (Parsons 2004, 187).
question what the good is can only be answered when we know the answer to the question what rightly loving is. A similar point can be made with respect to truth. We call something true, if the acknowledgement or rejection that is directed to it is right, Brentano says. So, what is rightly acknowledging or rejecting? In general, What is right judging?
First, we need to compare two experiences, the experience of a blind judgement with the experience of an immediately evident judgement. According to Brentano, the immediately evident judgement, that is, the self-evident judgement, is perspicuous (einleuchtend); the act of judging is characterised by clarity (Klarheit). The immediately evident judgement cannot and need not be grounded by other judgements made (idem, §26). Our judgement may be self- evident, because our act of judgement is an act of insight – we understand that a body is extended, and thereby the judgement is self-evident; or our judging may be an act of inner perception – I may judge that I have a colour sensation, the moment I have such a sensation, and my judgement that I see something red is thus a self-evident judgement. If the judgement is derived by self-evident principles from self-evident truths, the act of judging is discursive, and the judgement is called mediately evident; these judgements are evident in the sense that they are grounded in other evident judgements.
In order to understand what a blind judgement is one has to contrast a blind act of judgement with an act of judgement that is clear and perspicuous, that is, one that results in an evident judgement. A blind judging is based on an instinctive urge: we are neither able to give an epistemic ground for these judgements, nor do we judge with clarity or insight (idem, § 26). A blind act of judging is thus to be understood as a non-perspicuous act of judging. In a blind judging, the judgement has been made without evidence.
Then, by comparing the experience of a perspicuous act of judging with the experience of a blind judging we can see that the first act has a characteristic that is absent in the case of the blind one. ‘Jeder erfährt den Unterschied zwischen der einen und anderen Urteilsweise in
sich; in dem Hinweis auf diese Erfahrung muβ, wie bei jedem Begriff, die letzte
Verdeutlichung bestehen.’ (idem). Experiencing the distinction between the two acts of judging is thus part of an elucidation of the concept of judging rightly, and thus of the concept of truth. The question remains, though, what precisely the relation is between evidence and truth; although blind judging never results in an evident judgement, the judgement may nevertheless be true.
In his letters to Marty and Kraus, Brentano develops an epistemic account of truth from 1906 on (cf. Brentano 1966, 174, 175, and idem, 207). Shortly before his death in 1917, he
develops the new definition of truth more fully. Just as red, evidence is a simple characteristic (ein einfaches Merkmal, Brentano 1915c, §7). Such a simple characteristic cannot be
explained as a complex of attributes. Just as we have to compare our sensations of red with our sensations of other colours, in order to grasp the simple characteristic of red, we have to compare our evident judgements with non-evident judgements, to grasp the characteristic of evidence. There is no need, Brentano adds, to explain evidence in terms of clear and distinct perception, as Descartes has done (Brentano 1915c, 141). When we hear a sound, and we are conscious that it consists of many parts, though we do not clearly distinguish between these parts, our judgement that we hear this sound is evident, although we do not have a clear perception of the sound. An inner perception need not be completely clear in order to be evident (Brentano 1930, 149).
The concept of judgemental truth can now be explained in terms of the evident
judgement. If one simply identified truth with evidence, truth would be a very narrow notion.
We would no longer be able to speak of judgemental truth in case of a blind judgement. We need a more general notion of correctness in order to be able to speak of a true judgement although it may in an epistemic sense not be as it should be. In the explanation of the notion
of truth, Brentano points to a crucial distinction. It concerns the distinction between a judgement as I experience my own judgement and a judgement I experience as made by others, called an opinion (Brentano 1966, 293, 294). First, one looks upon a judgement made by oneself, and one thus apprehends a judgement from a first-person perspective. Then, one looks upon a judgement someone else is making, that is, one look at judgement from a third- person perspective. The opinion might be evident to the other person, but I myself have no
access to the evidence of the judgement made by someone else. The question whether he judges with evidence is here irrelevant. The distinction between my own judgement and the opinion of someone else, that is, the distinction between a first- and a third-person perspective on judgement replaces Brentano’s former distinction between the judgement I judge with evidence and my own blind judgement. The way Brentano now elucidates the notion of truth gives him a criterion to establish the truth of all judgements or opinions. The blind judgement I made in the past is now considered from a third-person point of view, and can therefore be called an opinion, too. The replacement of the notion of blind judgement by the notion of opinion is thus a replacement of a narrower by a broader concept that covers both my own blind judgement and the judgements made by others, whether they judge with evidence or not.
I will come back to the distinction between the first- and the third-person perspective in the final section.
As Brentano explains in a letter to Kraus, dated March 21, 1916, in order to
understand this notion of truth, I need to have an intuition of a judgement that is evident to me; second, I need to have an intuition of someone else’s opinion (Meinung), which agrees with my evident judgement in object, quality (affirmative or negative), tense and modality (with necessity or without). When the two objects are compared, the evident judgement and the opinion agreeing with it, I have an intuition of the agreement between the two, which intuition forms the basis of the concept of judgemental truth. Thereby, I also have a criterion
to establish the truth of the opinion of others: the evidence of my own judgement is used to acknowledge the truth of judgements made by others (Brentano 1966, 293, 294). The definition of truth in terms of evidence is thus in principle able to provide a criterion for the truth of judgements made by others, whether they are evident or not.
Such a criterion is precisely what is lacking in the classical correspondence definition of truth (idem). The most important problem of the correspondence definition is that it is not able to provide a criterion to establish the truth of a judgement. For, in order to know that a judgement is true, we need to have knowledge of the object to which the judgement is supposed to correspond, which immediately brings us to an infinite regress, or a circle in the explanation. Remember, though, that this criticism was already formulated by Brentano in his lecture on truth from 1889. Brentano is still willing to use the term ‘correspondence’ in his new definition, but it concerns now a correspondence between the blind judgement and the evident judgement: they correspond in object, quality, tense and modality. It is a
correspondence that the evident judger is able to establish, so that the regress does not arise.
On this account, the correspondence between the two judgements is internal, based upon the identity of the quality, object, tense and modality of the two judgements. It is better, though, not to speak of ‘correspondence’ here, as there is no point of agreement here with the traditional correspondence definition of truth.
In a letter to Kraus from April 13, 2016, Brentano makes a distinction between two notions of rightness or correctness: logical and qualitative correctness. He calls someone else’s opinion that agrees with my evident judgement true (idem, 300), or qualitatively correct, because it agrees with the evident judgement in quality, that is, in its being an
affirmation or denial. If the judgement is evident, the judgement is logically correct (logisch richtig): it is as it should be in an epistemic sense. By logically correct Brentano thus means epistemically correct. At a few other places, Brentano writes that the term ‘true judgement’ is
ambiguous: in the original sense it means evident judgement; in a metaphorical sense (in übertragenem Sinne)6 a blind judgement that corresponds in content with the evident
judgement is also called true (Brentano 1925, 150; and Brentano 1952, 142). Apart from these terminological distinctions in the word ‘truth’, truth is eventually defined in terms of
evidence:
Truth belongs to the judgement of a rightly judging agent, that is, of a judger who judges in the way someone would judge who judges with evidence; who asserts what the evident judger would assert.7
The evidence of the judgement is not a criterion of its truth – there is no need for a criterion if one’s judgement is immediately evident (Brentano 1966, 301): one sees that it is true.
Evidence is simply identified with truth in its original, strict sense; and, evidence conceptually involves truth in its derived sense. There is an internal relation between the notion of evidence and that of truth in its derived sense, as it is used in the definition given above: truth is
explained in terms of judgemental evidence. By comparing the evidence of my judgement I have a criterion for establishing the truth of judgements made by others (idem, 301, 302; cf.
Brentano 1915b, 133). There is no other way to establish and justify judgemental truth than by comparing it to the evidence of my own judgements.
6 Brentano would call this an ‘Analogie zum gleichen Terminus’ (pros hen analogy) following here what Aristotle says of the term ‘healthy’. There is a focal meaning of ‘truth’, when it means evidence, and there is a meaning derived from this focal meaning, when we call a blind judgement or opinion ‘correct’ insofar as it is related to the evident judgement.
7 ‘Es läuft dies alles eigentlich auf nichts anderes hinaus als darauf, dass die Wahrheit dem Urteile des richtig Urteilenden zukommt, d.h. (dem Urteile) dessen, der urteilt, wie derjenige darüber urteilen würde, der mit Evidenz sein Urteil fällt; also der das behauptet, was auch der evident Urteilende behaupten würde.’ (Brentano 1915a, 139). In his later writings, Brentano acknowledges nothing but objects and persons in his ontology; he is
What is judgemental evidence, according to Brentano? ‘[W]as evident ist, ist es für den, der das evidente Urteil fällt.’ (Brentano 1966, 290). Not only the evidence of the judgement that I am now thinking is phenomenologically accessible in this sense, it also holds for the evidence of the axioms: one has to have the content of the axiom clearly before one’s mind, in order to judge it with perspicuity, and I experience thereby the evidence of my judgement. Equally, I experience the evidence of the mediately evident judgements in the sense that I judge perspicuously that the judgement follows from other evident judgements made. It does not make sense to speak of Evidenz an sich, for this would presuppose a Platonic realm of bearers of truth and evidence (Brentano 1930, 157). Evidence is, for Brentano, an inner characteristic of our judgements.
‘It is evident to me says the same as it is certain to me.’ (‘Est ist mir evident, sagt so viel als, es its mir sicher.’ Brentano 1930, 144; cf. Brentano 1930, 156, and 1889b, 67). The identification of evidence with certainty needs some caution, for certainty is an ambiguous notion. In the quote above, Brentano uses ‘certain’ in the Cartesian sense of absolute certainty; no mistake is possible. I come back to this point. At other passages, he contrasts certainty with evidence. Whereas evidence has no degrees, certainty as subjective conviction comes in degrees (Brentano 1966, 290). Because evidence is internally related to truth, and because there are no degrees of truth, on Brentano’s account, Brentano has to understand evidence as involving no degrees. Furthermore, we may be subjectively certain of non-evident judgements; people tend to be very certain about their prejudices and beliefs obtained by habit. Absence of doubt is not defining for evidence, either (Brentano 1930, 144). Nor is the immediacy of our judgement defining for immediate evidence. The memory of past
experiences is often judged immediately with a trusting belief or faith (zuversichtlicher Glaube, Brentano 1930, 148), but these judgements are not evident.
Does the fact that evidence is phenomenologically accessible imply a form of
psychologism? Evidence is not a natural urge or force, according to Brentano (1915c, 141).
Our habits may be very strong, and we seem to be forced to judge in accordance with them, but they need not be true. The fact that the evidence of my judgement is caused by certain states of my brain, and can thus be described in accordance with natural laws, is irrelevant, for a blind judgement is equally caused by certain brain states in accordance with natural laws.
The psychologist confounds the evidence of the judgement with a result of natural processes in accordance with natural laws generally holding for a certain species (Brentano 1930, 125;
cf. idem, 157, and Brentano 1966, 254).
Evidence is not a feeling of necessity (Brentano 1889b, 66, note 27), but an inner characteristic of certain judgements: in case of the evidence of our inner perceptions, there is a clarity that the mental act, being the object of inner perception, exists. Brentano admits that not all aspects of the mental act need be clearly perceived (Brentano 1930, 149), as was explained above. In the case of a priori judgements, the inner characteristic of the judgement is due to an act of insight: by grasping the concepts involved, one is able to make the
judgement; the judgement is made evident by an act of insight (Brentano 1889b, 66).
Evidence is not a psychological, but a logical and epistemic notion. According to Brentano, if my judgement is evident to me, no one can judge the opposite with evidence. In this sense, the evident judgement has general validity (Allgemeingültigkeit, Brentano 1889b, 67, note 27).
In order to be able to defend the internal relation between truth and evidence, Brentano understands evidence to be infallible: evidence is real evidence. What is really evident should be distinguished from what merely appears to be evident; scheinbare Evidenz is opposed to wirkliche Evidenz (Brentano 1915a, 137). Evidence is infallible: ‘Error is excluded’ (‘Bei Evidenz ist Irrtum ausgeschlossen.’ Brentano 1930, 144). Is Brentano entitled to claim that
evidence is phenomenologically accessible and, at the same time, that evidence excludes error?
2. Questions and Criticisms
In his definition of truth presented above Brentano introduces the notion of someone who asserts what the evident judger would assert. Who is the evident judger he is speaking about?
If truth is explained in terms of evidence, and evidence has to be the evidence of my own judgement right at this moment, truth seems applicable only to a few things. It is for this reason that Brentano understands truth as what can be made evident, where evidence includes both immediate and mediate evidence. So, if it is (mediately) evident to me that 54+23=77, I am entitled to say that an evident judger would judge that way, for I am at such a moment an evident judger. The moment the arithmetical judgement is evident to oneself, one is entitled to say that any judgement with the same content made in the past is true, because one now knows that there could be such an evident judger in the past. This leaves us with quite a number of judgements for which we have no way to say whether it or its opposite would be judged with evidence by an evident judger. So, either Brentano has to admit that we are not entitled to say that all judgements are true or false, or he has to admit that what the evident judger would assert is beyond our cognitive grasp in quite a number of cases. If we take Brentano’s criticism of the correspondence definition seriously – that it does not provide a criterion to establish truth, we have to conclude that the first interpretation suits Brentano’s position. This means that Brentano can defend only the negative, weaker version of the law of excluded middle, as I have argued elsewhere (Schaar 2002/2003).
There are arithmetical theses that have never been made evident until now, while we do want to be able to call them true. Constructivists in mathematics, who also defend an epistemic account of truth, say that these mathematical theses are true insofar as we have a method to decide about their truth. As soon as we have a method of enumeration, we know that the results of counting in accordance with the method are true, if we have made no
mistakes in the application thereof. Brentano’s account of evidence does not exclude this form of idealisation, although he does not mention it.
In his definition of truth, Brentano says that a judgement is true, if it is judged in the way the evident judger would judge (urteilen würde, Brentano 1915a, 139; see §1). One may thus wonder to what extent an idealisation of the evident judger is allowed. Are we entitled to say that the evident judger in the definition can be identified with God? At several places, Brentano presupposes that God is a supreme knower. 8 God not only judges everything he judges with evidence, but also judges everything there is to know, for example, that I am thinking at this moment. Does Brentano defend the thesis that a judgement is true, if and only if it agrees with the judgement of God in aspect of quality, object and tense? 9 Brentano does not mention God when he gives his definition of truth. As we do not have any epistemic access to God’s judgements, it would bring in a form of idealisation that makes the definition sensitive to the criticism Brentano has formulated against the traditional correspondence definition of truth. Such a definition gives us no criterion to establish whether our judgements are true. It thus seems unlikely that Brentano allows for this form of idealisation, although it is not inconsistent with the definition he has put forward.
8‘daβ alle Wahrheit von Gott apodiktisch erkannt werde, wuβte schon Aristoteles,’ (Brentano 1966, 275; letter to Kraus, January 30, 1915).
9 I leave out here the point that one has to judge with the same modality (with necessity or without), for God
In the introduction I mentioned Nelson’s criticism of the use of evidence in epistemology.
Nelson argues that evidence cannot function as a criterion to distinguish knowledge from other judgements made. According to Nelson, either the concept of evidence involves that of truth, or evidence is nothing but an experience in our consciousness that can psychologically be established. (I) If evidence involves truth, it is impossible to decide whether a judgement is evident, for we may always be mistaken with respect to the evidence of our judgement; we are not able to distinguish true from merely apparent evidence. (II) If evidence is a
phenomenological notion, if evidence is given to us in experience, it is impossible to decide
whether an evident judgement is true; the evidence of the judgement is given to us, and precisely for this reason cannot conceptually imply truth. (Nelson 1908, 124).10 The problem of evidence cannot be solved, according to Nelson.
For Brentano, evidence is not a criterion of knowledge; evidence is rather defining for knowledge. Knowledge is precisely the evident judgement, where its evidence may be
mediate or immediate. I return to this identification of knowledge and evident judgement below in section 3. Nelson’s criticism does apply, though, to Brentano’s account of evidence, for evidence does involve truth on this account; at the same time, Brentano argues that I experience the evidence of my judgement: I know it when I judge with evidence; evidence is thus also a phenomenal character of certain judgements. Nelson argues that the same notion of evidence cannot involve both elements. One may argue against Nelson’s thesis that in some very clear cases, as in 1 + 0 = 1, and the judgement that I am thinking at this moment, the judgement is evident in the sense that it implies truth, as well as evident in the sense that the judging agent has a special experience of evidence. Nelson would answer that in the past people have often had the experience of evidence, while the judgement later turned out not to
10 Günther Patzig has given a similar criticism with respect to Husserl’s notion of evidence in the Logical
be evident. We have no means to distinguish real from apparent evidence, according to Nelson.
Nelson has a point. Brentano assumes that we are infallible in our judgements when we judge with evidence, and that we experience the evidence of our judgement at the same time. But, as Nelson would say, in case I experience my judgement as evident, error is not excluded. In order to grasp the judgeable content that 1 + 0 = 1, one needs to understand the concepts of zero, of one, and of numerical identity, and one has to understand how the latter concept applies to numbers. On all levels, a mistake is possible. The same point may be raised concerning the I think. The evident judgement that I am thinking involves the concept of the first-person (I) and the concept of thinking (think), and I may be mistaken in applying these concepts to the particular phenomenon experienced. This does not imply, though, that evidence has no role to play in epistemology.
Should we then distinguish two notions of evidence: an objective notion related to truth, and a subjective notion related to our consciousness? This will not help much: the objective notion of evidence will turn out to be an empty, non-applicable concept. As we have no access to objective evidence, it adds nothing to our analysis of truth. The criticism
Brentano formulated against the correspondence definition of truth, - it does not give us a criterion to establish whether a judgement is true -, applies to a theory that explains truth in terms of objective evidence. A subjective notion of evidence, which is unrelated to truth, will not help, either, for we loose then the relation between evidence and insight. Subjective evidence seems to be nothing but a psychological feeling of certainty. An account of evidence in which it turns out to be merely a psychological feeling cannot have any epistemic
relevance. Certainty as a psychological feeling accompanies blind judgements as well. In contrast to judgemental evidence, this feeling of certainty allows for degrees. With respect to some complicated axioms, our feeling of certainty may be of a low degree.
Self-evidence is an important notion in epistemology. For example, it directs us to the moment of insight in the concepts of zero, one, identity, and plus, through which we are able to judge that 1 + 0 = 1. The justification for this judgement is not to be found in other
judgements made, it is rather to be found in the insight obtained when we understand the judgeable content, and this is why we call the judgement (self-)evident. In our epistemology, we need both an account of how judgements are related to each other, and how the whole edifice of judgements made may be related to individual judgements, in order to be able to give an epistemic value to each of the judgement complexes that we know to make up a coherent whole.
The distinction Brentano makes between real evidence (wirkliche Evidenz) and sham evidence (scheinbare Evidenz) cannot help us to give an independent elucidation of the notion of evidence, for the distinction presupposes that the notion of evidence is already in use. The modifying term ‘sham’ applies to a noun ‘evidence’ whose meaning must already be
understood independent of the phrase ‘sham evidence’. And, the term ‘real evidence’ is primarily used in a negative way: ‘The judgement turned out to be not really evident’. Such a sentence makes sense only if the judgement was first held evident. These terms cannot
properly apply to judgement in a first-person perspective; they apply only to judgements made by me in the past, or to judgements made by others. We first need to understand what
evidence is in a first-person perspective.
At the height of the influence of logical positivism, when their criticism of the notion of evidence was largely accepted by the philosophical community, Wolfgang Stegmüller published a chapter on the problem of evidence in his Metaphysik, Wissenschaft, Skepsis (1954). At first sight, he seems to endorse Nelson’s criticism: the problem of evidence cannot be solved (Stegmüller 1954, 96) than Nelson’s answer. Stegmüller’s answer to the problem is entirely different, though. According to Stegmüller, evidence plays a role in all our
judgements. When we consider an argument to be valid because of the validity of Modus Ponens, we have to make the Modus Ponens rule evident, and we have to see the argument as
a special case of Modus Ponens. To someone who would deny that the argument is valid because of this rule, one can only say: ‘Don’t you see that that this is a case of Modus Ponens?’ In this sense, every argument makes use of evidence or insight. As soon as one
judges, Stegmüller argues, one relies on evidence, on insight. There is no way, though, to prove that there is self-evidence, for proofs can no longer be given when we speak of self- evidence. The question whether there is evidence or insight is absolutely undecidable
(Stegmüller 1954, 102). For, all arguments for evidence involve a vicious circle - they already have to use the notion of evidence, and all arguments against evidence contain a self-
contradiction - while arguing they presuppose that there is evidence. In order to be able to make any judgement, Stegmüller says, we need a pre-rational primary decision, a belief in evidence (Glaube an Evidenz). We have to assume that there is evidence or insight.
Stegmüller’s aim is not to understand what evidence precisely is; at several places he seems to speak of evidence in the sense in which Brentano uses it, evidence as real evidence, contrasted with merely apparent evidence.11 As we have seen, there is a notion of evidence prior in the conceptual order that is presupposed by our use of the term ‘real, infallible evidence’.
Epistemology and logic need a notion of evidence that is not identical to the notion of infallible, real evidence.
3. Truth and Evidence from a First-Person Point of View
In analytic philosophy the question what knowledge is, is answered by means of the question under what conditions we are entitled to attribute knowledge to a third person. Under what condition is the sentence ‘John knows that S’ true? The basic answer to this question is that John knows that S is true, if and only if S is true, John believes that S, and John is justified in believing that S. ‘John believes that S’ may be explained here as ‘John judges that S the moment he is thinking of S.’ It is often claimed that there are problems with the notion of justification in this definition, because it gives rise to Gettier cases. John may be able to give a justification for S, but if this justification has only an accidental relation to the truth of S, John’s belief is justified and true, while our intuitions say that it is not knowledge.
Could a Gettier case be formulated for Brentano’s account of truth and evidence? One important difference between Brentano’s account and the modern account of knowledge is that for Brentano knowledge is the right judgement in a strict sense (Brentano 1952, 2);
knowledge is evident judgement. We thus see that the term ‘true’ is absent in the definition.
As the evidence of the judgement implies its truth, there is no need to add the term. In fact, this means that our notion of judgemental truth captures the epistemic role of truth. As we have seen above, this internal relation between evidence and truth has been criticised, and I come back to this point at the end of the paper. For Brentano’s account of knowledge one cannot create a Gettier case, because there is an internal relation between the notion of evidence and that of truth. In this sense the notion of evidence is stricter than the notion of justification as we now use it. Apart from this difference, how is Brentano’s notion of evidence related to the modern notion of justification?
In general, modern philosophers do not call the immediately evident judgement justified, because it is assumed that a judgement is justified only if it is grounded by other judgements made. For Brentano, though, the evidence of the judgement makes the judgement
justified, also when the judgement is immediately evident.12 On a modern account, axioms are often understood as non-epistemic starting points of a system. Brentano, though, adheres to the traditional account of what an axiom is: what we use as an axiom should be known. When I have fully understood its meaning, I judge the axiom with evidence, and I thereby know the axiom to be true. The act of understanding makes the judgement justified, on Brentano’s account. In this sense Brentano’s notion of evident judgement is more inclusive than the modern notion of justified true belief.
The most important difference, though, between Brentano’s account of knowledge and the modern account consists in the fact that Brentano does not take knowledge attributions as starting-point for his analysis. One thing that is often unnoticed is that the knowledge
attribution approach presupposes an understanding of knowledge from a first-person point of view: there is a judging and knowing agent presupposed, who attributes knowledge to someone else. It is precisely this aspect of knowledge that Brentano is giving an account of.
For Brentano, the idea that knowledge is primarily a first-person notion is directly related to the thesis that we can understand a simple concept, later to be called a characteristic, only if a special case of the concept is given to us in an intuition (Anschauung). One can understand what knowledge is, only if one has or has had an intuition of a perspicuous act of judging, for knowledge is nothing but the evident judgement. Furthermore, the evidence of a judgement can only be established by the judging agent. Each judging agent has to make up his mind concerning the question whether the judgement is evident. Mediate evidence of one’s judgement can more easily be brought into the public sphere, as one can always point to the judgements made on which one grounds one’s mediately evident judgement. Nevertheless, such a process must come to an end, and then each judger has to see for himself whether the judgement is (self-)evident. Besides, the judging agent has to see for himself that the
12 ‘Berechtigt sind [unsere Urteile] nur dann, wenn sie uns einleuchten.’ (Brentano 1915c, 140). The context
conclusion follows from former judgements made. Immediate evidence is involved in the act of inference, for example, when we see the act as a special case of Modus Ponens.
As we have seen in section 1, for Brentano, in order to understand the notion of truth, we have to have had at least two intuitions: an intuition of one’s own perspicuous judging, and the intuition of a judgement of someone else, whose judgement agrees in object, tense, quality and modality with my own judgement. When I see the common characteristic, I have an intuition of a judgement that is true, and I am thereby able to grasp the notion of truth. The understanding of truth thus presupposes a distinction between judgement from a first-person perspective, when the agent looks at his own perspicuous act, and judgement from a third- person perspective, when the agent looks at a judgement made by someone else. Brentano acknowledges this with a terminological distinction: one’s own evident judgement is called a
‘judgement’, but the judgement of someone else, of which we do not know whether it is evident, is an ‘opinion’ (Meinung), a term used in contrast with knowledge. We thus see that whereas evidence is a simple notion, to be understood from a first-person perspective alone, truth is a complex notion, on Brentano’s account. This is confirmed by the definition of truth explained in section 1.
When are we entitled to say that a judgement is true, on Brentano’s account? If it concerns one’s own judgement, one can only establish its truth by making it evident. If it concerns the judgement of someone else, if it concerns an attribution, such as ‘John has a true opinion’, the only means we have is to compare John’s judgement with the judgement I make myself, and to see for myself whether it agrees in content with my evident judgement
(Brentano 1915b, 133, §5, and idem, 135, 136, §9). One is entitled to attribute a true opinion to someone else only if the judgement is evident to oneself. Attributions of true opinions are thus a derived phenomenon on Brentano’s account. One should not think that a comparison is needed in order to establish whether one’s own judgement is true. The central role of the first-
person perspective in Brentano’s account of truth and knowledge also extends to his logic, as the basic laws of logic need to be made evident to the judging agent. I have written elsewhere on the crucial role of the first-person perspective in logic, and will not go into it here.
This may all be an interesting alternative to the standard view on knowledge and logic, but it can only have philosophical value, if a crucial problem is solved. Evidence cannot be both phenomenologically accessible and involve truth at the same time. Can we keep the important insights explained above, while solving the problem? As Stegmüller has pointed out, the problem of evidence is unsolvable. We may follow him by saying that we need judgemental evidence in order to be able to judge at all. Such a notion of evidence, though, gives us no guarantee for truth. It is always possible to revise our judgement. This also means that logic and epistemology can no longer be called foundationalist in the traditional sense. If we find that our axioms lead to a contradiction, we know that at least one of our axioms, or one of our inference rules, cannot be right, and therefore cannot be evident. In this sense, evidence is related to truth. Only at this moment we are entitled to deploy the term ‘truly evident’. We may come to understand that the so-called axiom that seemed to be evident, is now no longer considered as ‘truly evident’. The term can only be used, though, because a prior term ‘evidence’ is already in use, standing for a notion of evidence as corrigible. We are often entitled to claim that our judgement is evident, but we are never entitled to hold that it is truly, infallibly evident.
References
Brentano, F. 1889a, ‘Über den Begriff der Wahrheit’ in Brentano 1930, 3-29.
--- 1889b, Vom Ursprung sittlicher Erkenntnis, Hamburg: Felix Meiner, 1969.
--- 1915a, ‘Über den Sinn des Satzes: veritas est adaequatio rei et intellectus <5. März 1915>’ in Brentano 1930, 137-139.
--- 1915b, ‘Über den Sinn des Satzes: veritas est adaequatio rei et intellectus <11. Mai 1915>’ in Brentano 1930, 131-136.
--- 1915c, ‘Gedankengang zur Lehre von der Evidenz <8. Juli 1915>’ in Brentano 1930, 140-143.
--- 1925, Versuch über die Erkenntnis, A. Kastil (ed.), Hamburg: Felix Meiner, 19702. --- 1930, Wahrheit und Evidenz, O. Kraus (ed.), Hamburg: Felix Meiner, 1974.
--- 1952, Grundlegung und Aufbau der Ethik, F. Mayer-Hillebrand (ed.), Bern: Francke.
--- 1966, Die Abkehr vom Nicht-Realen, F. Mayer-Hillebrand (ed.), Bern: Francke.
Chrudzimski, A. 2001, Intentionalitätstheorie beim frühen Brentano, Dordrecht: Kluwer.
Parsons, Ch. 2004, ‘Brentano on Judgment and Truth’, The Cambridge Companion to Brentano, D.
Jacquette (ed.), Cambridge: Cambridge University Press, 168-196.
Patzig, G. 1977 ‘Husserl on Truth and Evidence’ in J.N. Mohanty (ed.) Readings on Edmund Husserl’s Logical Investigations. The Hague: Martinus Nijhoff, 179-196.
Schaar, M. van der 2002/2003 ‘Brentano on Logic, Truth and Evidence’, Brentano Studien, 10, 119- 150.
Stegmüller, W. 1954, Metaphysik, Wissenschaft, Skepsis, Frankfurt am Main, Vienna, Humboldt Verlag.
