Descriptive Empiricism. Stumpf on Sensation and Presentation

RICCARDO MARTINELLI Descriptive Empiricism: Stumpf on Sensation and Presentation 1. Stumpf ’s empiricism The first book of David Hume’s Treatise of Human Nature opens with the following words: All the perceptions of the human mind resolve themselves into two distinct kinds, which I shall call IMPRESSIONS and IDEAS. The difference betwixt these consists in the degrees of force and liveliness, with which they strike upon the mind, and make their way into our thought or consciousness. (Hume (1739), 1) Hume believes that “it will not be very necessary to employ many words in explaining this distinction”. Notwithstanding some remarkable exceptions, e.g. in dreaming or madness, the difference between the two species of human perception is quite clear. Impressions exhibit a typical “force and violence”: ideas, on the other hand, are the “faint images” of impressions, as they occur “in thinking and reasoning”. Hume’s entire empirical reconstruction and classification of mental phenomena is based on this distinction. Impressions and ideas, from his point of view, differ essentially in the specific degree of “force and liveliness”. As this example suggests, the relationship between sense impressions and ideas necessarily represent a capital issue for any kind of philosophical empiricism. At least, this is certainly the case for Carl Stumpf. In the first place, Stumpf ’s philosophy is actually a peculiar form of empiricism, supporting his excellent experimental work in psychology. Along with a few analytical a priori logical truths, Stumpf recognised and appreciated synthetical a posteriori judgments as the main source of human knowledge (Stumpf (1939), 205). Secondly, this empiricism is bound to the assumption of a continuous gradual difference in intensity between “sensations” (Empfindungen) and “presentations” (Vorstellungen). Except for the lexical choice, Stumpf appears to agree completely with the beginning of Hume’s Treatise. Brentano Studien 10 QGHGLWLRQ2006), 81-100 82 However, Stumpf ’s empiricism does not differ only terminologically from that of the great Scot ancestor. Following Franz Brentano, Stumpf rejects any “genetic” (i.e. physiological or psychophysical) approach to mental phenomena in favour of a “descriptive” attitude. Thus, he consciously neglects the problem of the origin of ideas, focusing attention on mental phenomenology. At the same time, Stumpf did not simply reproduce his teacher’s Psychognosie. He permanently shared with Brentano the belief that philosophy follows no other method than that of natural science. Yet, Stumpf had a drastically different idea of what scientific methodology actually is. Unlike Brentano, he left no place for any prominence of inner perception and its alleged evidence. Accordingly, his picture of psychical life ends by revealing a rather different landscape. If compared with Brentano’s separation of sensory and noetic conscience, Stumpf ’s thesis of continuity between sensation and presentation confers a distinguished empirical character to his theory. More precisely, the demonstration that continuity between sensation and presentation is of an intensive kind, sharply contrasts with Brentano’s ultimate achievement that no presentation possesses intensity at all (Brentano (1911), 151-152).1 Halfway between Hume and Brentano, Stumpf confronts us with an innovative and highly developed solution for all of these classical problems. For this reason, and in this sense, his philosophy may be labelled as a descriptive empiricism. Unfortunately, Stumpf pronounced his last word on the problem of sensation and presentation only in 1918, in an essay for the proceedings of the Berlin Scientific Academy. Despite the promising title Empfindung und Vorstellung, two circumstances conspired to limit the circulation of the essay. Two years before his retirement, the septuagenarian professor was indeed mostly considered as a psychologist. The unpublished volumes of his Erkenntnislehre bear us witness that Stumpf was at the time neither uninterested in philosophy, nor too old for the task. Since it answers a difficult and fundamental question, the 1918 essay represents a very important step within the context of his late philosophical activity. Our study in the problem of sensation and presentation follows to some extent Stumpf ’s own development in this matter: § 2 presents his general views on psychology; § 3 provides a discussion of the doctrine of tonal fusion; § 4 analyses and discusses the theses of Empfindung und Vorstellung. Finally, in § 5 some conclusions are offered, with special attention for the problem of visual perception. 83 2. Sensation and judgment Stumpf ’s philosophy may be properly described from the starting point of the most general theoretical dichotomy he recognises; i.e., that of psychical functions and appearances. It is on this basis that Stumpf proceeds to his classification of the sciences (Stumpf (1907a), passim). According to him, natural sciences aim at a description of appearances (Erscheinungen), whereas human sciences concentrate on a description of psychical functions (psychische Funktionen). Physics deserves a central position in the former group, psychology in the latter. Besides sciences, Stumpf also allows for pre-sciences – i.e. phenomenology, eidology, and theory of relations – and for a post-science, metaphysics. Philosophy has a phenomenological preliminary task and a post-scientific metaphysical function. In both cases, philosophy and psychology are closely related to each other (although not identical), insofar as psychology is the fundamental Geisteswissenschaft. The dichotomy of Erscheinungen and psychische Funktionen explicitly emerges in Stumpf ’s philosophical work around 1906, but it is quite easy to show its previous traces and its remote Brentanian origin. Since 1874, Brentano had proposed a similar distinction between psychical act and intentional object: for example, the act of hearing is different from sound as its object (Brentano (1874), 111). Accordingly, Brentano distinguished two forms of “perception”; i.e., “inner” and “outer”. For him, inner perception is theoretically pre-eminent because of its possessing the unique character of evidence. Despite his general dependence on Brentanian psychology, Stumpf always kept far from such conclusions. Let us first notice that he deeply innovated Brentano’s terminology, and certainly not by chance. The word “appearance” (Erscheinung) immediately reminds us of Kant’s Critique of pure reason and of the neo-Kantian philosophy contemporary to Stumpf, while the reference to psychical functions (Funktionen) is quite typical for the opposite school of Fechner’s psychophysics. A short analysis of Stumpf ’s attitude towards these two radically conflicting traditions helps to understand why Stumpf did reform the basic lexicon of descriptive psychology. In 1891, Stumpf had taken an active part in the ongoing polemic between professional psychologists and neo-Kantian philosophers, prefiguring the later Psychologismusstreit. In an essay entitled Psychologie und Erkenntnistheorie, he established a sort of “independent compatibility” for philosophical reflection and psychological research. According to this principle, one can admit the results of transcendental reflection only whenever they do not conflict with scientifically ascertained psychologi- 84 cal truths (Stumpf (1891), 18). For example, it was rather easy for Stumpf to underline the inadequacy of Kant’s doctrine of space, which had already appeared untenable to F.A. Lange, Helmholtz and other pioneers of neoKantianism. Those thinkers struggled for rescuing the ethical and theoretical core of Kant’s philosophy from a potential defeat imposed by the impressive development of nineteenth-century physiology and psychology. Operating in this sense, neo-Kantian philosophers often saw themselves opposed to psychologists. In turn, despite an active commitment in philosophy, men like Fechner or Wundt were proudly unwilling to allow the superimposition of aprioristic philosophy on their experimental researches. Within this context, Stumpf ’s tenet that psychology cannot ever completely substitute Erkenntnistheorie contributed to mitigate the tones – and probably granted him the access to the Berlin University. The following passage expresses his typical style in such polemic debates: “We let now critical philosophers and psychologists shoot one another with their bullets. Instead of this, for the thing’s sake, we carefully sharpen the edge of our arguments […]” (Stumpf (1891), 4). Stumpf believes that the same scholar should separately cultivate, as he himself did, both disciplines: philosophy and psychology.2 Stumpf defends psychology, yet does not present himself as an enemy of Kant, thus supporting conciliation with his followers. From this perspective, one can easily explain the allowance for the concept of Erscheinung in his system. Stumpf does not think at any thing-in-itself as the counterpart for appearance. Philosophy is not allowed to assume arbitrarily any element a priori, but psychology on its side cannot claim more than a honest description of the given. Any other hypothesis must be drawn and defended at the later and different stage of metaphysics, regarded as a post-science (Nachwissenschaft) hanging on the most general laws of the psychical and physical reality. Stumpf generally legitimates metaphysical inquiries, but did not personally pursue them too intensively during his active career. He preferred working as “an expert among experts”; i.e., a scientist, a psychologist. Accordingly, he spent much of his time and efforts in the analysis of “psychical functions” and their laws. In 1860, Fechner published his influential Elemente der Psychophysik. The new science, psychophysics, was defined as the doctrine of “functional- or dependence-relations between Body and Soul or, more generally, between material and spiritual, physical and psychical world” (Fechner (1860), 8). The requirement of a functional relationship between the two worlds clearly appears from this quotation. Fechner introduces a mathematical, rather than metaphysical treatment for the problem of 85 relating physical and psychical realities. If one conceives of sensation – he claims – as a function of the stimulus, one could finally be able to find a mathematical expression providing the value of the variable ‘sensation’ for any stimulus-input. In short, Fechner pretends to imitate the successful strategy of Newton’s gravitation law: substituting philosophical sophistries with a general mathematical function, namely the famous logarithmic „Maßformel“ for the measurement of sensations. As his lexical choice reveals, Stumpf was actually tempted by a “functionalism” of this kind. Nonetheless, his mathematical psychology keeps remarkably distant from Fechner’s psychophysics. Sensations alone, for Stumpf, cannot solve the problem of making psychology a science. Besides them, the field of psychical functions entails other capital elements, namely judgments (Urteile). In Stumpf ’s view, judgments do not only belong to the intellectual sphere: sensibility itself presents an abundance of “sense-judgments” (Sinnesurteile), deeply rooted in any perceptual item. A definition and a wide treatment of sense-judgments are available in the two volumes of Stumpf ’s Tonpsychologie (1883 and 1890). Despite its title, and as Stumpf himself repeatedly claimed, this is actually a book in philosophy. The Tonpsychologie is indeed a living manifesto of descriptive empiricism, and perhaps the deepest and most representative among Stumpf ’s works. The first volume opens with a definition of sense-judgments: “When we determine a sensation as the note A, or as the Third of F, we express a sense-judgment; i.e., a judgment directed to sensorial appearances and raised by them” (Stumpf (1883), 3). The utmost importance of sense-judgments for Stumpf ’s theory clearly appears from this passage. After the first stages in their individual evolution, normal adults always let a sense-judgment go hand in hand with their sensations (Stumpf (1883), 7). Pure sensations alone, then, do not allow a sound psychological quantification: this is Fechner’s capital error. Recognising a multiplicity, an increase, a similarity, a “fusion” -or any other relation within sensibilitynecessarily involves sense-judgments. All this makes any psychological function much more complicated than Fechner conceded. Epistemological statements about psychology must account for these elementary facts, providing a thoroughgoing theory of Sinnesurteile. Stumpf recognises a specific “threshold of judgment” (Urteilsschwelle): only sensations falling above this threshold are noticeable, while others exert only indirect effects on perception (Stumpf (1883), 33-34)3. In Stumpf ’s theory, then, a “measuring doctrine of judgments” (messende Urteilslehre) explicitly substitutes psychophysics. Unlike Fechner, Stumpf regards sense-judgments as the starting point of any psychological measurement. However, how to measure judgments? 86 Stumpf is quite original in facing this capital question. First, he insists on the fallibility of sense-judgments. As a musician, he often experienced on others, and on himself too, some typical case of confusion. Unskilled or musically-ignorant people, but occasionally even experts, often mistake two heard tones for one, or higher tones for lower ones, and so on. As a scientist and an empiricist, Stumpf ascribes a great and positive value to this fact. Judgments are not only and not much useful for science insofar as they are true. On the contrary, the fallibility of perceptual experience reveals a circumstance of the greatest epistemological value. For Stumpf, judgments generally possess a certain degree of reliability (Zuverlässigkeit). Unlike Brentano’s evidence, reliability is a measurable quantity because it admits of different degrees: We do not talk about the evidence that the judgment has for the person who pronounces it. It is a particular question, yet one especially concerning logicians, that of establishing whether a sense-judgment, which has not been deduced from general premises but rather raised from appearances themselves, could ever possess that peculiar evidence one usually ascribes to universal logical axioms, making any proof superfluous and any doubt impossible for the person concerned. For the following researches it is indifferent, whether the judging subject ascribes such evidence to his judgment or not. It is only a matter of the degree of trustworthiness that such a judgment possesses for another person.4 Here Stumpf clearly rejects and theoretically overtakes Brentano’s philosophical methodology. The role of Brentanian evidence in granting the ultimate truth of scientific judgments is now covered by psychological experiment, bringing about a quantification of the degree of judgments’ intersubjective reliability. In his later work, Stumpf discusses the problem of inner perception again. He concedes that the assessment of truths about sense data requires particular (namely experimental) care; but this holds a fortiori for inner perception. No surprise that orthodox Brentanians will fiercely criticise this aspect of his thought (Stumpf (1939), 217; Kastil (1948), 199-200). In the first volume of the Tonpsychologie, Stumpf evaluates judgments about temporarily delayed tones, concentrating his analysis on pitch (or “tone quality”) and intensity. He demonstrates, for example, that Weber’s law applies to variations in intensity but not in pitch, as Fechner wrongly believed (Stumpf (1883), 299, 354). His most fruitful and problematic results, however, come only in the second volume of the Tonpsychologie, devoted to the analysis of simultaneous tones. 87 3. A theory of multiplicity We have already quoted the very first lines of the Tonpsychologie. Immediately later, Stumpf introduces the problem of sensation and presentation: Sensory appearances can be given, rather than in sensation, also in mere presentation (imagination, memory) and then nevertheless be judged: even a merely presented tone is occasionally recognised as A, or as the Third of another tone, and we name this judgment a sense-judgment too. In what follows, however, unless otherwise stated, we simply think of judgments concerning sensations.5 Far from giving a dogmatic, non-descriptive definition of sensation and presentation, Stumpf remarks here that sense-judgments occur in the same way when one thinks or remembers sensory situations. This implies an actual continuity between sensations and presentations. For the most part, experimental settings obviously make use of sensations, yet Stumpf shows himself confident that his scientific results would hold even for the corresponding presentations. In other words, sense-judgments follow the same general laws in both cases, being more or less reliable according to specifically different situations. Again, recalling Stumpf ’s experience as a musician helps to understand his views. As the last quoted passage clearly states, one currently asserts that a certain note is an A, or the Third of F, in silent thinking as well as in actual hearing. Musicians often compare a remembered tone with an actually heard one; they judge the former higher or equal to the latter, and so on. Tuning a bow instrument, for instance, implies abilities of this kind, quite ordinary for a skilled player like Stumpf. Musicians also imagine melodic voices when they silently read a score before playing it, and so does the artist during composition. However, as we already know, in the second volume of the Tonpsychologie Stumpf does not deal anymore with melodically successive tones. Simultaneous tones seem to require particular care from the perspective of presentation: can we imagine a chord in the same way we think of a melody?6 The problems involved by this task appear far more serious than in the previous case. As a general preliminary step, Stumpf individuates three different theoretical frameworks. If one admits that “many sensations at the same time, or just one sensation, or many sensations one after another” are involved, then a multiplicity-, unity-, or a contrast-hypothesis prevails (Stumpf (1890), 12). Accordingly, a chord is conceived as many tone-sensations fused together, or as a single chord-sensation, or finally as the result of an enormously fast alternation of tones in the mind. Stumpf definitely 88 embraces the first option. For him, the hearing or thinking of a chord implies the apprehension of an actual multiplicity of tones, contained in the appearance. The sense-judgment sometimes clearly distinguishes this multiplicity. More frequently, however, it recognises only partially its nature, or even completely misses the actual state of affairs. Such confusion does not mainly proceeds from subjective awkwardness or inexperience. Stumpf ’s experimental researches demonstrate that judgments concerning tonal multiplicity follow general rules, depending on which tones are chosen to build up the chord. Each different combination of tones differently influences the sense-judgment. In other words, any different tonal multiplicity is essentially characterised by a certain degree of “tonal fusion” (Tonverschmelzung), that is, by the strength of the tendency to form a sensorial whole (Empfindungsganze), rather than a mere sum of sensations (Stumpf (1890), 64, 128). Not to mention its Aristotelian roots, the modern origin of this idea goes back to Ernst Heinrich Weber’s experiments concerning the sense of touch. Weber had noted that, in weighing two bodies, successive evaluation considerably improves human sensitivity. On the contrary, if both hands simultaneously compare two weights, the two resulting sensations tend to mix, and then to confound the subject (Weber (1851), 85-86). This is precisely Stumpf ’s idea of tonal fusion. The simultaneous presence of two appearances deceives hearing and touch analogously. Variations in the reliability of judgments about hearing, however, are quite regular and liable to precise experimental treatment. For this reason, Stumpf typically explains hearing and musical phenomena on the ground of a detailed analysis of simultaneity and multiplicity. “Successive sensations -he writesform, as sensations, a mere sum; simultaneous ones, already as sensations, a whole”. His distance from theories like that of Ehrenfels’ „Über Gestaltqualitäten”, also published in 1890, appears far from any doubt.7 Now we can go back to the problem of presenting a chord in memory or imagination. In his study of the “laws of fusion”, Stumpf explicitly asserts what follows: “Fusion remains preserved even in mere fantastic presentation. When I merely present a C and a G as simultaneously resonating, I can present them only as fusing together, and precisely with the same degree of fusion they possess in real hearing”. (Stumpf (1890), 138-139) Not any feature of sensation extends to presentation. For instance, one always hears C and C-sharp together with the so-called “beats”, a waveeffect due to similarity in frequency. Yet, one can easily imagine those two tones without beats, or present them together with beats of an arbitrarily different speed –what is impossible in actual hearing. Hence descends for 89 Stumpf the need of establishing the principle, “not a priori expectable”, that the degrees of fusion possessed by remembered or imagined tones correspond to those of the namely, actually heard tones. Why did Stumpf worry about this point, insisting on the circumstance that tonal fusion persists in the case of remembered or imagined tones? In his view, tonal fusion has fundamental importance for the whole theory of music. Consonance and dissonance are defined in terms of tonal fusion degrees: the more two given tones are fused, the more consonant they appear. Would fusion not extend to presentations, no read or imagined (and then artistically composed) music could exist (Stumpf (1890), 333; see also Stumpf (1898) and (1911)). One must consider the challenge thereby launched by Stumpf to Hermann Helmholtz, who explained consonance and dissonance on the basis of purely physical factors analysed by the inner ear, such as beats and higher harmonic tones (Helmholtz (1863), 181183). What would then happen –asks Stumpf- when we merely present a piece of music? It would be absurd to believe that the peripheral nervous apparatus vibrates without external stimulation. However, at least in Helmholtz’s Lehre von den Tonempfindungen, any other hypothesis about musical and mental reproduction remains unexpressed. Should one accept, without any evidence, a process similar to the obscure “unconscious inferences” used by Helmholtz in explaining the process of space-construction? Stumpf ’s doctrine of tonal fusion solves all of these problems. The cause of fusion lies in the central, not in the peripheral nervous system. A “specific synergy” is elicited in the brain as soon as two stimuli are simultaneously present, no matter if they correspond to actually heard or merely imagined tones. Nevertheless, tonal fusion raises other no less serious difficulties. Theodor Lipps was probably the first to notice that, according to Stumpf ’s ideas, successive tones could never be consonant or dissonant (Lipps (1892), 562). In fact, successive tones cannot fuse together: by definition they form a “mere sum”. Lipps’ criticism heavily remarks how Stumpf focused all attention on multiplicity and simultaneity, disregarding the melodic factor. As a reply, Stumpf asserts that two physically subsequent tones should have an instant of “psychological simultaneity” in order to enter the same melody. During that little instant, “the second tone, which is heard, fuses with the first, which is still presented” (Stumpf (1898), 57). In short, Stumpf subscribes Brentano’s theory of psychological temporality. The mnestic trace of the “first sensation” (Brentano’s socalled “Proterästhese”) remains in mind until a second sensation comes to conscience, and so on. If true, this argument would actually refute Lipps’ 90 objection. For Stumpf, melodic perception succeeds thanks to psychological simultaneousness of tonal sensations and presentations.8 Stumpf ’s doctrine of tonal fusion then requires a very strong condition concerning the relationship between sensations and presentations. Not only the degree of fusion between two given tone-sensations is preserved when both tones are merely presented; moreover, Stumpf asserts that fusion also occurs between a sensation and a presentation.9 Therefore, it seems impossible to think of sensation and presentation in terms of two detached realms, the latter being a projection or a reproduction of the former. How should one think of them? 4. Sensation and presentation On the physiological level, Stumpf can easily justify fusion between a sensation and a presentation: the process originates in the brain from simultaneous recurrence of two heterogeneous stimuli. This may be an adequate “genetical” framework, but the question still waits for a satisfactory treatment on the descriptive level. In his Erscheinungen und psychische Funktionen, Stumpf had already proposed a solution. There, he defined sensations and presentations in terms of appearances belonging to a “first”, and a “second order” (Stumpf (1906), 4).10 Nevertheless, how is this “order” to be understood, and what does really means Stumpf with this? The essay Empfindung und Vorstellung finally offers a satisfactory solution for this question. Stumpf begins with distinguishing two main groups of theories from each other, according as one admits a difference in degree, or a specific difference between sensation and presentation. He adopts the former option: sensation and presentations are different degrees on the same scale of strength, or intensity. Once more, Stumpf chooses tonal appearances as favourite examples. They are ductile to the will of the experimenter, and liable to a precise quantification of their pitch and intensity. Moreover, if compared to sight or touch, hearing appears much less compromised with the problem of spatial determination, which is a permanent source of confusion (Stumpf (1918), 7-8). Thus limiting the field, Stumpf sets the general principles. Presentations are not wholly deprived of intensity, yet their strength is “extraordinarily smaller” than that of sensations. In the unitary scale of tonal intensities, the entire gamut of presentations falls below the group of sensations. No defined cut takes place between the two series: appearances falling in the intermediary zone are experienced either as sensations or as presentations. Precisely the frequent issues of confusion or 91 uncertainty between the two cases, for Stumpf, provide evidence for a theory of this kind. In sum, sensations and presentations can be distinguished in the sense that e.g. “a presented fortissimo is fainter than a heard pianissimo”(Stumpf (1918), 27). According to Stumpf, the range of presentation-intensities is less extended than that of sensations; however, he does not assert any direct projection of the one series upon the other. Rather, human soul originally disposes of a unitary continuum, within which the upper zone refers to sense-phenomena, the lower to presentation activities. One could try to render graphically this point, in the following way: a b P c M d S The arrow represents the auditory phenomenal intensities oriented from left to right; P, S, and M are the zones of presentations, sensations, and the middle uncertainty area; a, b, c and d, are intensity-values. In this way, the real continuity of presentations and sensations clearly results. Of course, the role of sense-judgments always ought to be properly appreciated. In fact, notwithstanding b<c, b is judged as “loud” (in presentation, i.e. as contrasted to a), while c is judged as relatively “faint” (in sensation, i.e. as contrasted to d). In the middle area M, the sense-judgment may waver between the two possibilities, or wrongly ascribe the appearance to S (e.g. in auditory hallucinations). So far mnestic presentations are concerned, Stumpf asseverates a general mathematical regularity: “strength-ratios remain preserved in memory, while strength-differences appear considerably reduced, on a small scale”(Stumpf (1918), 27).11 This means that, whenever d/c=b/a, the sense-judgment remarks that the two crescendo are “the same”. Accordingly, hearing a crescendo (e.g. from c to d), involves making use of a broader range of intensities than needed for presenting the corresponding crescendo from a to b. Stumpf poses no further requirement: particularly, he does not assume any special relationship between single items, like c and a, or d and b. In other words, presentation do not obey the Humean condition that any (simple) idea is always the reproduction of a previous sense-impression. It ought to be noted, however, that a remarkable exception for Hume concerns the filling of a continuous scale within sensory ideas. A single shade, e.g. of the colour blue, can be inferred by someone who has never seen it, insofar as all – or most - of the remaining shades are known to him. However, this appears to Hume a “particular and singular” instance, “scarce worth our observing” (Hume (1739), 6). On the contra- 92 ry, Stumpf takes this example seriously. For him, so to speak, this Humean exception acts as a main rule of continuity between the two realms. Stumpf ’s doctrine differs from classical empiricism for another no less important aspect. Just as no causal relationship between sensations and presentations is required, no causal linkage between “external” objects and sensations is explicitly postulated. This latter aspect deserves special attention. Stumpf is obviously aware that a definition of sensation and presentation could involve objects that are external to the mind, acting on it through sense organs. Nevertheless, as a descriptive psychologist he assumes the perspective of conscience itself. The “absence” of an external objects, therefore, would mean that the subject has a perception, and is (somehow otherwise) simultaneously conscious of the outer non-existence of the perceived object. He would then perceive and simultaneously pronounce a negative existential judgment (not a sense-judgment); only in this case he would actually have a presentation. However, such an intellectual concern is quite unsatisfactory for Stumpf. This does not imply that conscience is completely “spontaneous” in perceiving, i.e. wholly reality-indifferent. Stumpf is by no means an ingenuous idealist. Rather, his theory is properly justified on an evolutionary level: Intuitive sense-appearances of a certain species form, so far as their strength is concerned, a continuous series from the weakest to the strongest. In this series, separately taken and purely considered from the point of view of appearances, there is no fixed division point separating two classes from each other. That such a separation slowly develops within conscience during one’s growth has to do with the distinction of the body from its environment, and with knowledge that appearances belonging to a certain upper strengthzone usually arise by the effect of external objects and processes on our sense-organs.12 Knowledge that very strong appearances usually correspond to physical external causes is obviously acquired by experience. After this passage, Stumpf still insists in distinguishing the tasks of ontogenetic separation between the Self and the world, as contrasted with those of scientific classification. It is worth quoting this entire passage: For clarity’s sake in this matter, it is quite necessary to keep well in mind the relationship between the mark of intensity and that of the conscious relation to an external stimulus. We wholly recognise the significance of the latter for ontogenetic development: the rise of the entire distinction between sensation and mere presentation has to be reduced to that factor. Thereby is not yet said that it would be 93 conclusive for scientific classification. […] Once the minimal appearance-intensity corresponding to an external stimulus is established as the lower limit of the upper zone, all appearances whose intensity exceeds this threshold must be counted among sensations, even though the conscious relation to an external stimulus is either missing or replaced by the conscious relation to an internal one. Facts concerning threshold serve only for setting up the scale of intensities. They play a role, which is similar to that of the water’s freezing temperature in fixing a point within the continuous expansion of mercury. However, their service ends with this. The thermometer of intensities, set up in this way, has now become an autonomous tool of measurement, by virtue of which we perform the attribution of an appearance to the group of upper or lower intensities.13 Any reference to external objects in performing the distinction of sensation and presentation is thus inadmissible: the claim for a direct acquaintance with the physical conditions of experience deprives descriptive psychology of any scientific validity and meaning. Individual conscience is neither a stimulus-response mechanism, nor is it wholly indifferent to external references. Human development involves a repeated adjustment of the scale of consciousness; and the only possible criterion for this task is given by experience. 5. Visual perception Some interesting difficulties with Stumpf ’s doctrine have been raised by Paul Hofmann, a Berlin philosopher influenced by Wilhelm Dilthey and Benno Erdmann, in an essay also bearing the title Empfindung und Vorstellung (1919). If ever, notes Hofmann, Stumpf ’s theory may be accepted only for “intuitive” presentations, not for symbolic or abstract ones. “I believe – he writes – that Stumpf ’s exposition should be convincing for those who have for the most part concretely intuitive presentations. It would be convincing even for me, were it not my own way of presenting quite of another kind, namely essentially unintuitive” (Hofmann (1919), 5). To be sure, Stumpf would agree with this premise. He allows for a symbolic function of thought, but keeps it rigorously distinct from sensory presentations of the above mentioned kind: “Our thought actually operates, maybe for a larger part, with mere symbols and concepts. This fact, however, whose significance cannot be denied, should not be generalised as if there were no concretely intuitive presentational images at all”(Stumpf (1918), 9). On the contrary, Hofmann believes that intuitive 94 and non-intuitive elements are always interwoven – one could ironically say “fused” together – in one’s Erlebnisse. The tissue of living experience is made of both kinds of presentations; properly one should speak only of psychical acts being “for the most part” of sensible or abstract kind, according as the one or the other factor prevails. At any rate, Hofmann concludes, “the intuitive elements that are, so to speak, contained in presentational experiences, never reveal essential” (Hofmann (1919), 11). For this reason, he concludes, Stumpf ’s proposal fails. No presentation is purely intuitive; therefore, no presentation can be explained in terms of intensive degrees. This penetrating analysis includes also some remarkable mistakes. Unsurprisingly, issues of acoustic perception deceive Hofmann especially. Vivid auditory appearances such as the terrible cracking of grenades and the explosion of bombs “as I have heard so many times on the battle field”, he writes, are “presented without hearing anything in this” (Hofmann (1919), 6). In this dramatic recollection, written in 1919, Hofmann is not aware of giving the simple definition of auditory presentation (“without hearing anything in this”) rather than questioning its intuitive character. On the other side, he is right in observing that no kind of intensity applies to unintuitive – abstract or symbolic – manners of thought. It could be a matter of different standing points: Hofmann’s non-intuitive favourite modes of thought contrast with Stumpf ’s lengthy acquaintance with silent, yet neither symbolic nor abstract, auditory presentations. Nonetheless, in those years Stumpf had seriously considered the problem of visual perception too. In an essay entitled Die Attribute der Gesichtsempfindungen, he developed an original and lucid doctrine. Stumpf begins with a polemic against Ewald Hering. The famous Prague physiologist argued in 1874 that the impression of black does not arise from the absence of visual sensations, as if it was the zero of an intensity scale. Deep black is a “positive” sensation, as well as bright white. Hence, Hering says, one should admit of two distinct intensive series for black and white, which is manifestly absurd. The solution for this dilemma comes easily if one eliminates the concept of intensity from psychological analysis of vision, and then from general theory of perception (Hering (1874), 88). With this, Hering aimed at defeating Fechner’s psychophysics and its calculation formulas for the intensity of sensations. It is obvious enough, however, that his observations represent a serious threat for Stumpf ’s entire doctrine. In reply, Stumpf reminds the reader of the Kantian Antizipationen der Wahrnehmung, where a proof that every sensation possesses a degree of 95 “intensive quantity” is provided (Kant, Kritik der reinen Vernunft, B 207208). However, Stumpf cannot content himself with an a priori argument. Visual sensations, for him, do admit of intensity, though only in a relative sense. The relative amount of blue or red could be described as the relative intensity, or “partial strength” (Teilstärke) in the impression of violet. After all, language records expressions like “intensive blue” in a quite similar sense (Stumpf (1917), 44). This raises, in turn, the problem of the “strength of the whole”; i.e., violet in Stumpf ’s example. Once more, Stumpf solves the question by means of a multiplicity theory. We saw that a chord, for him, is rather an actual multiplicity than a unitary perceptual issue, as Gestaltists assert.14 Similarly, any compound colour really consists of parts endowed with a specific intensity – even though, in both cases, the “intensity of the whole” is not equal to the sum of the two compounding relative intensities. Again, for Stumpf this is a matter of sense-judgments, their failure, and the calculation of their intersubjective reliability. The Kantian problem of anticipating perception slides into the purely empirical question of establishing the conditions for a variation of intensity within each situation. In the visual field, intensity reveals a factor much close to constancy, therefore passing easily unobserved. Nevertheless, it is occasionally possible to isolate and vary a specifically visual intensity. A complete account of the discussion contained in Die Attribute der Gesichtsempfindungen, however, would take us too far afield. In any case, once a general (relative) visual intensity is admitted, this kind of theoretical difficulty disappears. The intensity of visual presentations, for Stumpf, lies below that of Augenschwarz, the natural black of a closed human eye (Stumpf (1918), 48). With this, our reconstruction of Stumpf ’s theory of sensation and presentation comes to an end. One can obviously subscribe to Hofmann’s criticism, and say that the concept of presentation is ambiguous; or either follow Hering in a total refusal of the concept of intensity, on which the entire theory is grounded. Various other reasons against this doctrine could also be found, but entering this level of discussion exceeds our present tasks. In conclusion, Stumpf takes up the great tradition of empiricism when he assumes a difference in degree between sensations and presentations. However, his continuity with classical empiricism is only partial. Stumpf himself was well aware of this fact. For him, the “old theory” of an intensive difference between presentation and sensation failed to underline the psychological continuity of the two series, thus overestimating the gap between them. Of course, Hume and other classical empiricists were chiefly worried about the Cartesian allowance for innate ideas, and 96 therefore insisted on the circumstance that any simple idea follows on a previous sensation. Yet, as Stumpf would put it, the problem of the origin of ideas is a problem in “genetic”, not in “descriptive” psychology. Descriptive psychology raises different philosophical problems, posing further requirements. In Stumpf ’s hands, the metaphysical question of the so-called “external” reality slides into an empirical problem. Experience calibrates the scale of intensities, thus providing adaptation of the Self to its environment. Accordingly, no preliminary metaphysically oriented definition of sensation and presentation is necessary. From this perspective, Stumpf ’s empirical development of descriptive psychology ceases to appear as a somehow eccentric variation upon Brentanism. His solution for this classical problem is rather one of the most coherent developments within that philosophical tradition. Literature: Brentano, Franz (1924). Psychologie vom empirischen Standpunkt. Bd. I. ed. by O. Kraus. Leipzig: Felix Meiner. Brentano, Franz (1907). Über Individuation, multiple Qualität und Intensität sinnlicher Erscheinungen. In Chisholm, R.M. & Fabian, R. (eds.). Untersuchungen zur Sinnespsychologie. Hamburg: Felix Meiner. Brentano, Franz (1925). Psychologie vom empirischen Standpunkt. Bd. II. ed. by O. Kraus. Leipzig: Felix Meiner. Fechner, Gustav Theodor (1860). Elemente der Psychophysik. 2 vols. Leipzig: Breitkopf & Härtel. Fechner, Gustav Theodor (1877). In Sachen der Psychophysik. Leipzig: Breitkopf & Härtel. Fechner, Gustav Theodor (1886). Über die psychischen Maßprincipien und das Weber’sche Gesetz. In Philosophische Studien 4, 161-230. Gelb, A. (1911). Theoretisches über Gestaltqualitäten. In Zeitschrift für Psychologie 58: 1-58. Helmholtz, Hermann (1877). Die Lehre von den Tonempfindungen, als physiologische Grundlage für die Theorie der Musik. Braunschweig: Vieweg. Helmholtz, Hermann (1954). On the Sensations of Tone, as a Physiological Basis for the Theory of Music. New York: Dover. Hering, E. (1874). Zur Lehre vom Lichtsinne. IV: Über die sogenannte Intensität der Lichtempfindung und über die Empfindung des Schwarzen. In Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften, mathematisch-naturwissenschaftliche Klasse. Wien: Hofmann, 85-104. Hering, E. (1919). Empfindung und Vorstellung. Ein Beitrag zur Klärung psychologischer Grundbegriffe. Kantstudien, Ergänzungshefte, n. 47. 97 Holenstein, E. (1972) Phänomenologie der Assoziation. Zu Struktur und Funktion eines Grundprinzips der passiven Genesis bei E. Husserl. Den Haag: Nijhoff. Hume, D. (1739) A Treatise of Human Nature. ed. by L.A. Selby-Bigge, Oxford: Oxford at the Clarendon Press, (1978) (with new materials and corrections). Kastil, A. (1948). Ein neuer Rettungsversuch der Evidenz der äusseren Wahrnehmung (kritische Bemerkungen zu Stumpfs Erkenntnislehre). In Zeitschrift für philosophische Forschung 3, 198-218. Koffka, K. (1917). Probleme der experimentelle Psychologie I. Die Unterschiedsschwelle. In Die Naturwissenschaften 5, 1-15, 23-28. Köhler, W. (1913). Über unbemerkte Empfindungen und Urteilstäuschungen. In Zeitschrift für Psychologie 66, 51-80. Lipps, Th. (1892). Der Begriff der Verschmelzung und damit Zusammenhängendes in Stumpfs Tonpsychologie, Band II. In Philosophische Monatshefte, 28, 547-591. Lotze, R.H. (1853). Psychologische Untersuchungen. In: Über die Stärke der Vorstellungen. In Zeitschrift für Philosophie und philosophische Kritik 22, repr. in Kleine Schriften, ed. by D. Peipers, Leipzig: Breitkopf & Härtel (1885-91), vol. 3, 72-99. Maltzew, C. von (1913). Das Erkennen sukzessiv gegebener musikalischer Intervalle in der äußeren Tonregionen. In Zeitschrift für Psychologie 64, 161-257. Nadel, S. (1927). Zur Psychologie des Konsonanzerlebens. In: Zeitschrift für Psychologie 101, 33-158. Rath, M. (1994). Der Psychologismusstreit in der deutschen Philosophie. München: Alber. Stumpf, C. (1883). Tonpsychologie. Erster Band, Leipzig: Hirzel (repr. Amsterdam: Bonset, 1965). Stumpf, C. (1890). Tonpsychologie. Zweiter Band, Leipzig: Hirzel (repr. Amsterdam: Bonset, 1965). Stumpf, C. (1891). Psychologie und Erkenntnistheorie. In Abhandlungen der Königlich Bayrischen Akademie der Wissenschaften, philosophisch-historische Klasse. München. Stumpf, C. (1898). Konsonanz und Dissonanz. In Beiträge zur Akustik und Musikwissenschaft. Leipzig: Barth. Stumpf, C. (1906). Erscheinungen und psychische Funktionen. In Abhandlungen der Preussischen Akademie der Wissenschaften, philosophisch-historische Klasse. Berlin. Stumpf, C. (1907). Die Wiedergeburt der Philosophie. In Philosophische Reden und Vorträge. Leipzig: Barth (1910), 161-196. Stumpf, C. (1907a). Zur Einteilung der Wissenschaften. In Abhandlungen der Preussischen Akademie der Wissenschaften, philosophisch-historische Klasse. Berlin. Stumpf, C. (1911) Konsonanz und Konkordanz. Nebst Bemerkungen über Wohlklang und Wohlgefälligkeit musikalischer Zusammenklänge. In Zeitschrift für Psychologie, 58, 321-355. 98 Stumpf, C. (1917). Die Attribute der Gesichtsempfindungen. In Abhandlungen der Königlich Preussischen Akademie der Wissenschaften, philosophisch-historische Klasse. Berlin. Stumpf, C. (1918). Empfindung und Vorstellung. In Abhandlungen der Königlich Preussischen Akademie der Wissenschaften, philosophisch-historische Klasse. Berlin. Stumpf, C. (1939) Erkenntnislehre. Erster Band, Leipzig: Barth. Weber, E.H. (1851). Die Lehre vom Tastsinne und Gemeingefühle, auf Versuche gegründet. Braunschweig: Vieweg. Notes: 1 Only sensations exhibit differences in intensity, interpreted by Brentano as real variations in the number of parts involved in each item. See Brentano (1896), passim. 2 The figure of a wide-cultured scholar, philosophically as well as scientifically minded, is embodied by Leibniz, and later by Lotze and Fechner. See Stumpf (1907), 178. This strictly personal conciliation of philosophy and psychology heavily limits Stumpf ’s solution for the Psychologismusstreit. See Rath (1994), 205. 3 Against this hypothesis see Köhler (1913), Koffka (1917). 4 „Wir sprechen also nicht von der Evidenz, welche das Urteil etwa für den Aussagenden selbst besitzt. Es ist eine besondere Frage, welche aber mehr den Logiker interessirt, ob jemals ein Sinnesurteil, welches nicht aus allgemeinen Prämissen abgeleitet sondern durch die betreffenden Erscheinungen selbst veranlasst ist, jene eigentümliche Evidenz besitzen könne, wie man sie den allgemeinen logischen Axiomen zuschreibt, die jeden Beweis überflüssig und jeden Zweifel für den Betreffenden unmöglich macht. Für die folgenden Untersuchungen ist es einerlei, ob der Aussagende seinen Urteil diese Evidenz beimisst. Es handelt sich nur darum, welcher Grad der Vertrauenswürdigkeit dasselbe für einen Anderen besitzt“. (Stumpf (1883), 22) 5 „Sinnliche Erscheinungen können statt in der Empfindung auch in blosser Vorstellung (Phantasie, Gedächtnis) gegeben sein und dann ebenfalls beurteilt werden – auch ein blos vorgestellter Ton wird unter Umständen als a, als Terz eines andern erkannt – und auch dies nennen wir ein Sinnesurteil, haben jedoch im Folgenden, wo nicht ausdrücklich anders bemerkt ist, nur die Beurteilung von Empfindungen im Auge“ (Stumpf (1883), 3). 6 Hermann Lotze, Stumpf ’s teacher in Göttingen, had already raised this question. He suggested that remembering a chord actually means having a very fast alternation of tones in mind. Lotze (1853), 83-84. Stumpf criticises this thesis, labelled as “contrast-hypothesis” (see below). 7 Stumpf (1890), 64. Stumpf kept silence on the problem of Gestalt qualities, or laconically dealt with it (e.g. in Stumpf (1906)). A deep criticism of Ehrenfels’ 99 perspective is available in a dissertation held under his guide and influence by Adhémar Gelb, who will later adhere to the Gestalt theory. By the way, Gelb discusses also Husserl’s “figurales Moment”, criticising it in strictly Stumpfian terms, and radically distinguishing it from tonal fusion. Gelb (1911), 50-51; for a theoretical discussion of the topic see Holenstein (1972), 124. 8 Stumpf ’s defence is obviously weak. Köhler’s assistant Catharina von Maltzew will later note that a different psychological effect emerges if the two tones are presented in ascending or in descending order. The degree of fusion cannot be the only, nor the main factor involved. Maltzew (1913), 190-192, 239; see Nadel (1927), 90-92. 9 E.H. Weber truly anticipates this point. “This mixing (Vermischung) is a very interesting fact […]”, he noted in a passage quoted by Stumpf ((1890), 61). Yet, Stumpf omits to quote what follows in Weber’s text: “[…] but even more interesting is the fact that a sensation, which is already past and only remembered, i.e. only presented by means of phantasy, can be related in the very same manner to a present sensation”. Weber (1851), 86. 10 Later he revised and conciliated this position with his new results: Stumpf (1918), 83, fn. 1. 11 Stumpf ’s discussion reminds here of a very interesting topic in the debate concerning psychophysics. Fechner carefully distinguished a Verhältnishypothese from an Unterschiedshypothese, according as one assumed the ratio or the difference between two particular sensations to be constant. The preliminary choice between these two hypotheses deeply influenced all final psychophysical formulas. Fechner embraced the latter alternative, while the former expresses the sense of Brentano’s criticism to psychophysics. See Fechner (1886), 174-175; Fechner (1877), 24-25. See also Brentano (1874), 97-99. On a different level, Stumpf follows Brentano in maintaining the constancy of intensive ratios, thus introducing an element of relativity in his doctrine. 12 „Die sinnlich-anschaulichen Erscheinungen einer bestimmten Gattung bilden ihrer Stärke nach eine stetige Reihe von den schwächsten bis zu den stärksten. In dieser Reihe gibt es an und für sich, rein erscheinungsmäßig betrachtet, keinen bestimmten Trennpunkt, der zwei Klassen voneinander schiede. Daß eine solche Scheidung im Bewußtsein des heranwachsenden Menschen sich allmählich vollzieht, hängt mit der Unterscheidung des eigenen Körpers von der Umgebung und mit der Erkenntnis zusammen, daß Erscheinungen, die einer gewissen oberen Stärkezone angehören, der Regel nach durch Einwirkung äußerer Objekte und Vorgänge auf unsere Sinnesorgane zustande kommen“. (Stumpf (1918), 81) 13 „Es ist zur Klarheit in dieser Sache durchaus notwendig, das Verhältnis zwischen dem Intensitätsmerkmal und dem der bewußten Beziehung auf einen äußeren Reiz genau im Auge zu behalten. Die Bedeutung des letzeren innerhalb der ontogenetischen Entwickelung erkennen wir vollkommen an: die Entstehung der ganzen Unterscheidung zwischen Empfindung und bloßer Vorstellung ist darauf zurückzuführen. Aber damit ist nicht gesagt, daß 100 es für die wissenschaftliche Klassifikation das entscheidende sein dürfte. […] Nachdem einmal die geringste einem äußeren Reiz entsprechende Erscheinungsintensität als untere Grenze der höheren Zone festgelegt ist, müssen alle Erscheinungen, deren Intensität diesen Punkt überschreitet, zu den Empfindungen gerechnet werden, auch wenn die bewußte Beziehung auf einen äußeren Reiz fehlt oder durch die Beziehung auf einen inneren ersetzt ist. Die Schwellentatsachen dienen nur zur Eichung der Intensitätsskala. Sie spielen eine ähnliche Rolle wie die Gefriertemperatur des Wassers für die Fixierung eines Punktes innerhalb der stetigen Ausdehnung des Quecksilbers. Aber damit ist ihre Leistung erschöpft. Das so geeichte Thermometer der Intensitäten ist nunmehr ein selbständiges Maßinstrument geworden, mit dem wir die Zuteilung einer Erscheinung zur Gruppe der oberen und unteren Intensitäten vollziehen“. (Stumpf (1918), 82) 14 Köhler’s conception of Gestalt as “unity of sensation” (Empfindungseinheit) is criticised and carefully distinguished from the author’s own position (Stumpf (1917), 57-58).