Almayev Nickolay

Part 1.The problem of meaning's description

The problem of musical meaning - does the music has some meaning at all, and if yes, then how it may be induced by it – leads us to the problem of how meaning should be described. In a little bit more naive formulation the same question may sound as what do we call “meaning”, or what do we mean saying “meaning”?

Temporary leaving aside that this term has a long history of application to various problems of psychology and philosophy, we can suppose that it is the meaning of a word that is typically associated with the word “meaning”. Example of a meaning may be found in a dictionary. It is a designation of that to what a word refers with the help of some other words.

If this implication is correct then speaking about the meaning of the music we shall very soon face some considerable difficulties.

First of all it is rather hard to say to what does music refer. Although Sloboda (1986) has found several examples of events that are mimicked by music (like the singing of birds, the sound of springs etc.) in general music refers to nothing in the outer world. This state of affairs immediately puzzles our typically extraverted attitude towards the human psyche, which is also shared by the most of contemporary psychology, and consists in definition of internal states of psyche through the external events of environment.

Music somehow immediately refers to psyche, without any particular references to external events.

Another disappointment is connected to the rather restricted ability of words to describe that intensive and complicate living–through processes that might be induced by music. Unlike the meaning of text it is almost impossible to say what this or that piece of music is about. Each listener can propose plenty of contents and typically none of them is interested in that, what the composer had in mind while creating this piece.

Nevertheless, it is obvious that natural language and music can bridge each other through the description of emotional states i.e. of how this or that content is experienced.

Elsewhere (Almayev N., 1999a) I've argued, that several problems of psychology that are connected to language and thought cannot be solved in an unified manner if we shall not find some entities of the fixation of meaning that are of a finer grain than the words. (For example, it is not clear what is in commons between the several meanings of the words, between metaphoric and literal sense, between some characteristics of a situation and usage of a specific words with the purpose of adequate description of this situation, and some more.)

The same and even in some greater extent refers to music, that is on the one hand definitely has no significant connection to the meanings of words, but on the other hand cannot be called meaningless.

Theoretical tools for meaning description were based on Husserlian concept of intentionality and inner temporality as the most primitive form of it.

Please refer to (Almayev N., 1999 a) and/or (Almayev N., in press) for the details of philosophical background and methodology. We found also that intentionality and inner temporality are not sufficient for the description of meaning realization in concrete living-through process. Particularly, Husserlian terms could not describe why some intentions might realize themselves in consciousness in spite of the ego's will, whiles the other can not do the same despite all the ego's desire.

Psychoanalytic concept of "psychic energy" that ascends to the works of Freud seemed to be rather promising although demanding clarification both from the phenomenological and natural science approaches.

It is obvious that "energy" is a dimension that is independent from the dimension of "objects," because any object in our psyche can acquire energy and lose it. At the same time, energy is not separated from intentions because when an object has energy its intentions are fulfillable, while they cease to be such when it loses energy. Thus, "energy" describes not the actual but the potential state of intentionality.

Energy is the ability to restore continuously the "whatness" of the former objects, but also the ability to create the "whatness" of the new objects and the new moments of inner time.

If we combine an "intentional" definition of energy with the "neural a priori" (Almayev N., in press, paragraph: "The Necessity of Intentions…") we will have to postulate the existence of a specific function that reserves a "resource of identification," or a "whatness resource," in order that any kind of object might exist in the psyche. Thus, psychic activity should be considered not only in the dimension of objects, but also in the independent dimension, that of energy management.

If we are allowed to use a computer metaphor, energy acquisition will stand for the process of memory allocation that is necessary for a program to work. The fact that our psyche is far from being similar to a computer of the von-Neumann architecture does not mean that it does not need some information resource in order to pass from one state to another. Energy may be defined as an information resource that is required to provide the loading of ever new neural units in order to connect with the active units. Correspondingly, the usage of such resource decreases when the connections are already well established and functioning.

In various psychoanalytic theories the concept of psychic energy is closely linked with the concepts of consciousness and unconsciousness, and, correspondingly, with those of "Ego" and "Id." Although psychoanalytic theories share this common point, they are substantially different in many other respects. In some theories the relation between the Ego and the Id is complicated by the participation of the "Self" which, e.g., for C.G. Jung, obviously means something different than for H. Kohut or D.W. Winnicott. Every such theory deserves an analysis that could easily exceed the volume of this paper by many times.

However, in order to start the present analysis, the following preliminary definition may be proposed that describes the relation between consciousness and the unconscious as:

between something that is made by us (i.e., refers to the ego and demands energy) and something that is made by itself (i.e., refers to the Id and may spare some energy).

Such definition presupposes a gradual relation between the different levels of implementation of activity, or the different states of formation of activity, and not a firm border between the entities. It corresponds to the basic ideas of the Russian physiologists as P.K. Anokhin(1978), N.A. Bernstein(1967), and P.V. Simonov (1985).

We can postulate the existence of at least two variables of psychic energy management that are necessary for the latter and obviously constitute some significant characteristics of all living-through processes. These are: 1) the intensity of energy demand (in which the ego is presently engaged); 2) the estimation of the level of energy (presently available to the ego).

These functions may be viewed as the dimensions of the space of emotional states. For example, in depression that is very often accompanied by anxiety the demand for energy is high (anxiety), while its actual level is low (depression). On the contrary, feeling joyful and free (I mean the kind of feeling that is best of all expressed by Luesher's (1983) yellow color, from the eight-color set) corresponds to the low demand for energy combined with a high level of available energy.

The state that corresponds to Luesher's red—directed activity—may be characterized as a high energy demand combined with an equally high level of available energy.

The empirically found dimensions of the semantic differential that are reproduced in almost all the factorizations without any considerable dependence on the material of scaling are: attitude (good-bad), strength (strong-week), and activity (active-passive) (Osgood Ch. et al,1957, Osgood Ch., 1976)

The strength factor easily corresponds to the possession of energy or its lack. The activity factor may be traced to an intensive search for energy, or the absence of the above. As for the first factor, its meaning may be interpreted as a protention of some energy resources for the sake of an object (good), or an inhibition of such protention (bad).

These examples are intended to show that the two aforementioned functions are easy to detect in the living-through processes of emotional states. At the same time, we can hardly hope to find them empirically without taking into consideration the intentional modifications of inner temporality. Moreover, an attempt to consider the intentional and energetic components of the living-through processes separately will lead to a simplification and a loss of the precision of our models.

Physically sound is a measured in frequency (Hz). But how can frequencies expressed in Hz produce meaning? The physical scale expressed in Hz should be somehow correlated, so to say, with the "human scale" which is at the same time physical and spiritual. Human body is a universal mediator between "nature" and "spirit".

Therefore, I've suggested that mechanism of resonance play key role in the process of meaning's induction that is conducted by music.

Sounds may possess symbolic meaning stemming from the association with the parts of body to which they are “tuned in” through resonance.

The unconscious is always considered as something that is "under" consciousness and its center: the ego. We can easily observe that the lower sounds resonate with the chest or even the stomach, while the higher with the throat and this or another section of the cranium. Throughout all of the Indo-European tradition beginning from the Vedas the chest (heart) is associated with the will and passions, while the stomach with drives or desires (unconscious and relatively more abound in energy), and the head—together with some sapient division—with reason or the mind (conscious but very often lacking energy).

Such observations, as well as the everyday experience of speech and music, lead us to formulate the basic assumption of the present model: the rising of the tune corresponds to the ego's more intensive demand for energy, while its falling to the ego's energy demand of reduced intensity.

The main principle of the inner temporality of consciousness that was discovered by Husserl as early as in 1917 (Husserl, 1969) is as follows: that which is impressed transforms into that which is retended and that which is retended passes to that which is protended. This basic passive transformation encounters the ever new impressions that enrich and correct it.

In many of his late works, and most of all in the last voluminous "Experience and Judgement" (1938), Husserl described the many levels of synthesis at which intentional structures experience modifications during the functioning of our consciousness. This multi-level character of synthesis is one of the basic principles of our model.

To all these characteristics, two independent dimensions must be added: first, the demand for energy, and, second, the estimation of the level of energy.

The “understanding” of a musical piece occurs when such a pattern of energy distribution is found that fits both the retended Gestalt of the melody and the living-through process that is meaningful for a personality.

Every content, in order for it to continue being active in the psyche, needs a certain amount of energy. This is precisely why we can consider every sound as demanding a certain amount of energy for its subsequent retention and protention. Correspondingly, the amount of sound in a particular period of time—i.e., the intensity of energy demand—is, perhaps, the most important characteristic of a living-through process. In traditional psychology and physiology its correlate is arousal. In music it is encoded in the tempo. The realization of concrete retentions and protentions of each tone substantially changes depending on the context of the tempo. Going of tune down corresponds to reduction in the ego's energy demand, although it depends on tempo to what a cause such a reduction may be attributed. It may be attributed either to the absence of energy due to the fact that additional energy is not available (see Almayev N. in press, example 2, in "Lento"). Or, on the contrary, - to a surplus of energy, when additional energy is not necessary (see ibid. example 1, in "Allegro molto").

The intensity of energy demand that is encoded by the pitch of the tone should not be confused with the intensity of energy demand that is encoded by the tempo. In general, we can express this relation as the one between "who" demands which is encoded in the pitch—it may be the "intellect," "will," "passions," and “how much” which is encoded in the tempo.

The following preliminary clarification may be proposed: ego's energy is the energy that might be further invested into any object whatever, it is, so to say, "pure" energy, i. e. allocated resource without given destination. Psychic states concerning management of the "pure" energy are prototypical for what is called intellectual activity. Object's energy is the same information resource although already allocated for a certain object. If ego begins fulfilling the objects intentions i.e. lets it into the consciousness, object shares its resource with the ego, and ego begins feeling more abound in energy, whiles at the same time more attached to the object, less "free". That is the state of "the passions". The state of "the will" may be understood as some combination of the ego's own energy demand with participation of energy that is being "stored" in some "valuable" objects.

The rhythmic organization of a musical piece may also be viewed from the standpoint of energy and inner temporality. Energy requires a certain time to be accumulated before its consumption. The regularity of such a process that combines the phases of accumulation and consumption—which permits to protend both the next consumption and the next accumulation—is encoded in the rhythm.

Everything we said before did not presuppose the existence of this or another musical system. That which is said about energy—to some extent including even the rhythm—refers to speech (its intonations and speed) as well as to music.

Now let us consider an example from the European diatonic system and see how our model can be applied to the explanation of the difference between the major and minor keys. In C-major the first, third, and fifth degrees are C-E-G, and in C-minor C-Eb-G. The psychological difference between these two sequences is obvious: the major key sounds much more confident, abundant in energy, while the minor key sounds obviously lacking in energy. Since the first and the last degrees of both sequences are equal, the psychological difference must be rooted in the different order of the intervals. According to our basic assumption, the rising of melody reflects a demand for energy.

Let energy be counted in Libs (from Libido, the name for psychic energy in the works of Freud and Jung), one Lib being equal to a semitone, 1Lib = 1 semitone (that is a rather preliminary and rough approximation).

Then, in the case of the major key we have a rise of first 4, and then 3 Libs, while in the case of the minor key it is first 3, and then 4.

Let us first consider the case of the major key. After the first tone (C) has finished sounding it is retended, and the protention of the same tone and the same energy demand is active. Then a higher tone (E) sounds, and correspondingly we have a higher demand for energy (+ 4). This rise in the energy demand is reflected by our consciousness at the second level of the synthesis and, in its turn, is being protended. It means that we passively expect the next rise in energy by the same 4 points. However, the next sound (G) requires only + 3. Consequently, at the third level of the synthesis our consciousness reflects a relative decrease in the energy demand and correspondingly we have a feeling of a surplus of energy: 4 - 3 = 1.

In the case of the minor key the situation is slightly different at the first and second levels of the synthesis, but is crucially different at the third. The transition from C to Eb requires only 3 Libs, but the further transition to G requires 4 Libs. Thus in the minor key we have a relative increase in the energy demand, instead of its relative decrease in the major. Correspondingly, 3 - 4 = -1.

Now we can easily feel the reason why, in particular in the 17th and 18th centuries, music was often considered as closely related to mathematics. (Mozart himself said that music is joy that wants to calculate itself).

The parallel to mathematics is obvious here: the relation that determines whether the key is major or minor corresponds, so to say, to the second derivative from the function of the rising of the tune.

In the two aforementioned examples (Almayev N., in press) we considered low energy demand in combination with the high level of available energy, and high energy demand in combination with low level of available energy. One can refer to the famous R.Wagner's "Flight of Walkure" theme as an example of intensive energy demand combined with the big amount of available energy.

Main assumption of all the proposed approach is that meaning of both natural language words and music might be described in one and the same way through the models of meaning realization. Being living-through processes meaning realizations are subjected to those intentional modifications, which were described by Husserl and to which the two functions of psychic energy management were added.

If that could be considered true, then the task of mutual elucidation of music and natural language becomes actual. We need natural language words for qualification of musical pieces, and also music from its' site being very well structured quantitative system could help in elucidation of how the meanings of the words are realized internally.

Therefore, first of all the reliability of correspondence between verbal qualifications and musical pieces should be checked. Will the subjects who share the common native language and belong to the same culture, but who are differently educated in general and in music in particular, show consistency in their qualification of different musical pieces?

In order to answer this question we (with my student Elkhimova L.- this work was for her graduating diploma at the Moscow State Open University) conducted in 1999 empirical research.

Two experts in music (teachers of a musical school) were asked to select 6 melodies (3 -19-th century Romantics, and 3 - "contemporary" Rock) and to describe each of them with the help of 4 adjectives, which will suite each composition best of all. Experts were also asked to have in mind, if possible, metaphors of taste.

As the result 24 descriptors were obtained. Some of them evidently implied estimation of tempo: "agile", "turbulent", the other were more or less resembling traditional adjectives from Osgoods (1976) semantic differential, several were very specific like: "aerated like Soda water", "little bit sweet" ("Land of confusion"), "strict" ("Du hast"). Correspondence of adjectives of different languages is the separate topic, therefore I will not try to translate the whole list.

Experts were asked only to propose metaphoric definitions for each composition, but not to compare compositions to each other according to the definitions that were already made.

Different subjects of the same culture but different educational and social background will be in general agree with the experts in describing the musical compositions by the adjectives of their native language (Russian).

H1. Subjects categorizations of different compositions will be significantly different and coincide with that of experts.

H0. Subjects categorization of different compositions will be random and will not coincide with experts estimations.

The following 6 compositions were selected: "Flight of a bumble-bee" by N. Rimsky-Korsakov, "Hungarian Dance" by J.Brahms, F.Kreisler's "The Torments of Love", Genesis "I can't dance", Genesis "Land of Confusion", Rammstein "Du hast".

20 Subjects, 10 male, 10 female predominantly young (17-35 years) with different musical preferences and educational background participated. Each of them received 6 blanks each blank contained all the 24 descriptors (one blank for one composition). They were asked to estimate correspondence of each composition to all the descriptors scaling from 1 - fits very poor to 7 - fits very well.

ANOVA with repeated measures was applied for the scores of each descriptor as a dependent variable and compositions as the levels of independent (grouping) variable. Distribution of all the dependent variables was quite close to normal. Data for each descriptor were computed separately. Compositions significantly differed according to all the descriptors. 9 cases can be considered as the "definite hits" - experts descriptions achieved the highest score within the subjects estimations and there was significant difference between the leading composition and the next one closest to it. In 7 more cases 2 compositions one of which was those one that was predicted by experts shared first place with no significant difference between them.

In the other 8 cases the composition that was predicted by experts was either among three or more leading compositions or differed significantly from the first one.

Nevertheless, in almost all the cases prototypical compositions were estimated significantly higher than the mean by corresponding descriptor.

We have also performed regression analysis of scores obtained for each descriptor as a dependent variable and tempo as an independent one.

Results were very different for the different descriptors. It was high for descriptors that evidently presupposed tempo estimation, low for all the taste-based metaphors and moderate for the rest. The greatest dependency on tempo R-sq.=0.661 (S- function) was detected for descriptor "Energetic". There were plenty of Quadratic and Cubic functions that served as the best approximations for various "U" and "inverted U" type dependencies of some descriptors on tempo. For example, in the case of descriptor "lucid" best approximation was cubic function with R-sq.=0.381, while linear approximation provided only about 15 % of explained variance.

In all the taste based metaphors descriptors R-sq. for the dependency on tempo was very small, just about several per cents.

In general, H1 may be considered as accepted and H0 as rejected. All the compositions varied significantly in all the descriptors. All the descriptors may be considered as adequate because the mean scores of corresponding compositions were significantly higher than the mean.

As for the fact that in 15 cases selected composition shared its primacy with one or more other compositions, it is rooted in the experimental task as it was formulated for the experts. They had to propose four descriptors for each composition, although not to evaluate which composition fitted this or that descriptor best of all. As the result descriptors with high loading on tempo has repeated several times for each composition. "Dynamic", "turbulent", "exciting"("agitating") for the "Bumble-bee", "agile" for "Land of confusion", "energetic" for "Du hast", "cheerful" for "I can't dance" and some other adequate translation of which puzzles me. As the result, compositions with the higher tempo either occupied the first rank or formed the whole group on the base of contrast to slow "Torments of Love", as it was the case with the descriptor "cheerful" that was supposed to designate "I can't dance".

The most striking and unexpected result for us was that the subjects reproduced the strange taste-based metaphors like "aerated" and "little bit sweet". Mean scores of the "Land of confusion" by those descriptors were significantly higher than that of any other composition. All my attempts to propose any reasonable explanation for this event ended as yet without any success.

"Strict" descriptor has also reproduced with the great difference of "Du hast" from all the other compositions. The latter may lead to formulating a hypothesis of how "strict" is encoded by music. "Strict" means that protentions of "not allowed" objects will be repressed quickly and without hesitations. "Du hast" differs from the other compositions by the substantially greater number of pauses which brake the melody. We always stay with some expectation of prolongation of previous sounds when the next pause happens. This explanation of course has to be tested in a special experiment that could help to identify concrete temporal and other characteristics of melody that are responsible for the encoding of "strictness".

The meaning of "Aired" (not to mix with "aerated like soda-water") which was initially proposed for "Land of confusion" seems to be encoded by the constant rise of tune without any significant dependency on the tempo. The first in this descriptor was the "Bumble-bee flight", although the second was slow "Torments of Love" that, nevertheless is characterized by the almost constant rise of tune.

The tempo although being very important category cannot predict the results solely. Even in the case of "Energetic" in which R-sq. was the biggest, the most energetic was "Du hast" with tempo=115, while not significantly different from it "Bumble-bee" had tempo=188!

Nevertheless, Regression analysis seems to be the most appropriate procedure for evaluation of influence of music variables on the estimation of this or that descriptor. Correlation (that presupposes linearity and is so widespread in the different branches of Psychology) is hardly may be of some value because most of the dependencies, as we have seen, are considerably different from the linear.

What is needed for an exploratory stage of investigations in relations between meaning of the words and meaning of musical pieces is to be able to include more significant variables into regression equations. Such variables might be: number and duration of rises and falls of melody, number and duration of pauses, "speed" of the rise and "speed" of the falls, etc. Unfortunately, we neither had software that could count such statistics for melodies, nor time to calculate it manually.

Concrete constants like, for example, the time at which estimation of available energy takes place is the matter of another type of experiments of a more precise character. We plan to apply schemes of one subject and paired comparisons of stimuli that will vary only in the duration of an event that is under investigation in order to determine the constants.

Almayev N. (1999a) Dynamic Theory of Meaning: New Opportunities for Cognitive Modeling. Web Journal of Formal, Computational & Cognitive Linguistics. http://fccl.ksu.ru/fcclroot.htm . http://fccl.ksu.ru/winter.99/cog_model/proceedings.htm

Almayev N. (in press). The Concept of Psychic Energy and the Phenomenon of Music. Analecta Husserliana: The yearbook of phenomenological research / Published under the authority of the World Institute for Phenomenological research and learning - Dordrecht: Kluwer Academic Publishers.

Penultimate draft is available at:

http://www.psychol.ras.ru/strukt/ALMAEV/penult.htm

Anochin P.K., (1978).Beitraege zur allgemeinen Theorie des funktionellen Systeme (Jena: Fischer,

Bernstein N.A. (1967). The Co-ordination and Regulation of Movements (Oxford: Pergamon Press,

Husserl E. (1939). Erfahrung und Urteil , Prag:Academia.

Husserl E. (1969). Gesammelte Werke Bd.10. Vorlesungen ueber inneren Zeitbewustssein. Den Haag. Martinus Nijhoff.

Luescher M., (1983). Luesher Color Test. London-Sydney: Pan.

Osgood Ch., Susi C.J., Tannenbaum P.H. (1957). The Measurment of Meaning rbana: University of Illinois Press.

Osgood Ch. E., (1976). Focus on Meaning vol.1.The Hague-Paris: Mouton.

Simonov P.V.(1985). The Science of the Human Higher Neural Activity and Artistic Creation. Moscow: Nauka.

Sloboda, J. (1986). The Musical Mind. Oxford: Clarendon Press.