Logos Multilingual Portal

6 - Reading - Part two

In the previous unit, we said that, in our mind, there has to be a sort of inner code, or subverbal code that, based on our perceptive experience, classifies possible perceptions (including perceptions of words) and subdivides them into cognitive types (CT) that are not words, but not entirely defined mental entities. Eco (1997) uses as an example the Aztecs and the horse: this animal, before the Spanish landing, was unknown to them, therefore alien to any cognitive types of their cultural heritage.

Nevertheless, where was, to an Aztec, the concept of horse, since he did not have it before the Spanish landing? Of course, after seeing some horses, the Aztecs must have created a morphological pattern not so different from a 3D model, and it is on this basis that we should infer the consistency of their perceptive acts. When I talk about CT, however, I do not mean just a sort of image, a set of morphological tracts or motorial features [...] That is to say, we can state that the CT of the horse from the beginning possessed a multimedia character 1.

At first, the cognitive type is something absolutely independent from the name of the object, or even the possibility to name it; it is something that only the person who has perceived the object in question can cause himself to recognize what he has perceived; therefore it is catalogued in a sort of inner subjective, idiomorphical code.

It was not necessary to name the object-horse to recognize it, in the same way as I can eventually feel a sensation inside me that is unpleasant, though indefinable, and understand just that it is the same I felt the day before 2.

In other words that we need to name something just if (and when) we behave as social animals and want, or have to, communicate with others. In contrast, in the autistic relation between my ego and itself, in order to reason about concepts or objects, I do not need any external language: neither the natural code made of words, nor other, artificial, codes. I nonetheless do need what Eco calls "cognitive type" (and Vygotskij "inner language") in order to identify a sensation or an object and classify them mentally so as to make more and more complex and differentiated the structure of my perceptive-cognitive apparatus.

ON THE NET - (english)
Routledge Encyclopedia of Translation Studies.


ON THE NET - (italian)
Riferimenti letterari 1
Riferimenti Letterari 2


  When we read a text or listen to a speech, as we held before, "we compare tokens to a type"3 and this process occurs at two levels, twice in a row.

  In a first phase, we put side by side the graphic/auditory token to a graphic/auditory type that is part of our repertoire of signs, of phonemes, or, more often, of recurring graphemic or phonemic patterns. This first phase allows us to make out the letter or word or locution or phrase ' or, whenever the given token doesn't match exactly any types found in our repertoire ' to find near matches of the tokens with the types present in the repertoire, and to choose among them the most plausible in the given context and co-text.

  Moreover, simultaneously, our mind analyzes the quality and the quantity of the discrepancy between token and type, and conjecture what sense we can attribute to such discrepancy: we therefore create also a (meta-) typology of variances from the type.

  If, for example, I find on paper the word "tonite" [sic] and in my repertoire I can't find this graphical pattern, but I find a near match with the word "tonight", I can infer that "tonite" may be a peculiar or local form of that word. Moreover, basing myself on previously registered deviations from the standard pattern (i.e. on my encyclopedic knowledge), I can infer that it is a word often found on signs outside restaurants and inns of a given area (because, let's suppose, I had previously met the graphical pattern "lite" and I had realized it was a local way of writing "light". From that experience, I had got to the standard deviance that, in my mind, matches the way of writing in a given geographical area).

  During the second phase, when we already have hold of the relation graphic token-graphic type, we have to enact a second token-type comparison in order to identify, based on the graphic type, the cognitive type evoked by the given graphic type. In other words, we have to pass from the phase when we "think of a word"4 to the phase in which we think about every meaning evoked by that word.

  The images evoked in one's mind by a given word do not match perfectly the ones evoked by that same word in the mind of any other speaker of that same natural code. The first limitation of inter-subjective communication lies therefore just in this rough match between the mental images corresponding to "horse" in the writer/speaker and the mental images linked to "horse" in the reader/listener. This happens because the subjective experiences (and the subjective images) linked to "horse" for the transmitter and for the receiver are not the same.

  The first loss produced in the act of verbal communication ' in the case of reading ' is caused by the subjectivity of the sign-sense match, due to the different individual experiences, to the idiomorphic nature of the relation of affective signification characterizing every speaker even within one natural code.

  «This requires processing at the syntactic, semantic and pragmatic levels [...]» 5.

  The mental processing of the read verbal material is of a syntactical nature when we try to reconstruct the possible structure of the sentence, i.e. the relations among its elements. In contrast, it is of a semantic nature when it identifies the relevant areas within the semantic field of any single word or sentence; and it is of a pragmatic nature when it deals with the logical match of the possible meanings to the general context and to the verbal co-text.

  Moreover, the text is analyzed in two ways:

[...] micro- and macro-analysis of the actual text: monitoring for cohesion and coherence, and checking for coherence between the actual text and the potential text-type of which it is a token realization [...] 6.

Microanalysis has the purpose of verifying text cohesion and inner cohesion of the single units of text among them. Macro-analysis is aimed at controlling coherence and cohesion between the created text and the category, the model to which the text refers. For example, if the text is an instruction booklet for a household appliance, or a story for a newspaper, often there are models for such types of text to which we frequently ' consciously or unconsciously ' adhere.

  The decoding of a message in the mind of the reader is a sort of compromise between these two kinds of analysis, because the bottom-up analysis, one semantic unit at a time, doesn't ever turn out the same results as the top-down analysis of the text as an entity that has its own coherent structure.

There is, in other words, a trade-off between the micro-/bottom-up analysis of the text at clause level and the macro-/top-down analysis of text as an entity 7.

As we can see, the reading of a natural code is not an aseptical or passive process of assimilation of universally definite concepts as it happens in a mathematical equation. Reading involves in itself cognitive differences and, consequently, interpretive differences. Even while we are reading, and the object of our perception are words and not things, we are led by cognitive types that help us catalogue the experience of possible writings, both in graphic and in semantic terms, in order to increase our perceptive-cognitive apparatus as readers, to speed up our decoding processes, to sharpen our critical capacity. The reader

may try to understand the meanings emanating from the text, or abandon himself to bizarre associations and free developments. I speak in terms of polarities, because no reading can prevent imagination to run free [...] 8.

The difference between a reader and a critic is negligible: the reader trying to understand has the same attitude as the critic, who is a systematic, methodical, self-aware reader. While reading

it is inescapable to compare two systems, the text system and the reader system; the critic action is substantially made up of such comparisons. 9

In the next unit, we will deal with mental processes linked to writing.


BELL R. T. Psycholinguistic/cognitive approaches. In Routledge Encyclopedia of Translation Studies. London, Routledge, 1998, p. 185-190. ISBN 0-415-09380-5.

ECO U. Kant e l'ornitorinco. Milano, Bompiani, 1997. ISBN 88-452-2868-1. English translation: Kant and the Platypus: Essays on Language and Cognition; translated from the Italian by Alastair McEwen, New York, Harcourt Brace, 2000.

SEGRE C. Avviamento all'analisi del testo letterario. Torino, Einaudi, 1985. ISBN 88-06-58735-8. English translation: Introduction to the Analysis of the Literary Text, with the collaboration of Tomaso Kemeny; translated from the Italian by John Meddemmen, Bloomington, Indiana University Press, 1988. ISBN 0253331064.

VYGOTSKY L. S. Myshlenie i rech´. Psihologicheskie issledovanija. Moskvà-Leningrad, Gosudarstvennoe social´no-èkonomicheskoe izdatel´stvo, 1934. English translation: Thought and Language; translated from the Russian and edited by Alex Kozulin, Cambridge (Massachusetts), MIT Press, 1986.

1 Eco 1997, p. 109.
2 Eco 1997, p. 111.
3 Eco 1997, p. 113.
4 We think of its sound, its shape on the paper, but we stop there, we don't go on decoding; it's something we all have experienced trying to read while our mind was busy with other thoughts: so we read the words as sounds, as graphic patterns, without getting any sense from them. "[...] the inner language has to be considered not as a language without sound, but as a very peculiar verbal function with an original structure and particular functioning modes which, just because it works in a totally different way from external language, in passing from one plane to the other one is connected with it in a dynamic, indissoluble way" (Vygotsky 1934).
5 Bell 1990, p. 187.
6 Bell 1990, p. 187.
7 Bell 1990, p. 187.
8 Segre 1985, p. 10-11.
9 Segre 1985, p. 11.