MENTAL
 Main Menu
 Comparisons
 MENTAL vs. Mentalese


MENTAL vs. mental
 MENTAL vs.
MENTALESE

Is MENTAL the language of thought?

"My view is that a language cannot be learned unless we already know one. The language of thought is known but not learned, that is, it is innate."(Jerry Fodor)

"The mind has a language of its own: Mentalese" (Steven Pinker).



The Universal Language of Thought

Fodor's Hypothesis

In 1975, Jerry Alan Fodor, in his work "Language of Thought", launched the hypothesis of the existence of a mental language, internal, innate and universal, common to the entire human species: the language of thought or Mentalese.

The characteristics of this hypothetical language would be, according to Fodor, the following: Fodor was one of the first philosophers to highlight the analogy between the human mind and the computer in advocating the computational theory of mind: the human mind can be considered a computer, since it contains internal symbols and an internal logic.

For Fodor, the goal of psychology should be to discover the "program" that constitutes the mind, not the physical aspects that enable that program to be executed. To discover the mind is to discover the software of the mind.


Fundamentals

The language of thought hypothesis is supported by 5 theses:
  1. Representational realism.
    Thought possesses a real system of representation, i.e., all mental content is expressed by explicit elements (tokens).

  2. Linguistic thinking.
    The representational system underlying thought is of a linguistic type, that is, it consists of syntactic elements (tokens) capable of expressing propositional meanings by means of the semantic compositionality of these syntactic elements.

  3. Distinctiveness.
    The language of thought is distinct from all natural spoken languages.

  4. Innatism.
    The language of thought is innate to all human beings and is unique: there is only one mental language for the entire human species. Fodor, defends a radical conceptual innatism: all concepts are innate.

  5. Semantic completeness.
    The language of thought is semantically complete, that is, it contains all the resources necessary for human beings to express or grasp any idea or concept.
The hypothesis of linguistic thought has the following characteristics:
  1. Productivity.
    Humans can produce a potentially infinite number of semantically distinct thoughts. The most obvious and simplest explanation is to assume that thought consists of linguistic elements (tokens) that combine to produce such an infinite variety of thoughts.

  2. Systematicity.
    There is a whole system of linguistic relations between the kinds of thoughts we are capable of producing. These relations are basically of a combinatorial type. For example, if we think "John loves Mary", we can also think "Mary loves John". And if we think "green circle" and "red square", we can also think "red circle and green square".

  3. Semantic compositionality.
    The meaning of a thought is based on the meaning of its component elements.

  4. Simplicity-complexity.
    Many thoughts are extremely complex semantically, and language is the only means we have for constructing them by combining simpler thoughts.

  5. Introspection.
    We are aware that we have thoughts that accompany us when we speak or when we pick up thoughts from others.

  6. Relationship inner world - outer world (the methodological argument).
    There is a causal relationship between inner world and outer world, between our thoughts and our actions. The fact that these actions occur in sequence and that they are of a rational type constitute arguments in favor of the existence of an inner language and of a logic associated with that language.
Therefore, if thought is of a linguistic type, there must be an internal language, a metalanguage, a mother language, a language of languages responsible for thought.


Pinker's position

Steven Pinker, in his 1994 book "The Language Instinct", includes a chapter (the third) devoted to Mentalese. He defines Mentalese as "The hypothetical language of thought, or representation of concepts and propositions in the brain in which ideas are formulated, including the meanings of words and concepts" [Pinker, 2009].

Almost 20 years after the hypothesis launched by Fodor, Pinker has reinforced and popularized this concept. His main ideas are: Pinker adduces 5 reasons for holding that thought cannot be natural language:
  1. The difficulty of expressing some ideas with words that do not exactly fit what one wishes to express.

  2. We remember the essence of what we have heard, but not the exact words.

  3. If thought depended on words, no new words could be coined.

  4. Without natural language one has thoughts (e.g., deaf people and babies), even thoughts of an abstract kind.

  5. Translation from one natural language to another would be impossible without Mentalese, the common and universal language of reference.
Pinker also provides 5 arguments in favor of Mentalese, citing existing problems in natural language:
  1. Ambiguity.
    Natural language is often ambiguous. But our understanding is not. Therefore, the instrument of our understanding (Mentalese) cannot be natural language.

  2. Lack of explicit logic.
    Logical relationships in natural language are often not explicit. This is related to the famous frame problem. [see Applications - Artificial Intelligence - The Frame Problem].

  3. The co-reference problem.
    In natural language it is not explicit how pronouns refer to their antecedents.

  4. Deixis.
    In linguistics, deixis refers to pointing to elements of discourse, such that often the context determines the meaning.

    Deixis is the part of semantics and pragmatics that is concerned with words that serve to indicate other elements. Words such as tú, hoy, aquí, esto, are deictic expressions, which serve to indicate people, situations, places, etc.

  5. Synonymy.
    Different words mean the same thing. Then there must be "something else" that is beyond those words.

Mentalese vs. Universal Grammar

Universal Grammar (GU) is a linguistic theory postulated by Chomsky, which holds that there are universal grammatical principles innate in all human beings and manifest in all natural languages.

For the ancient grammar of Port-Royal (1660), called "general grammar," the deep structure of all languages is the same, for it is a reflection of the forms of thought and of certain fundamental properties of the mind.

Also the linguist and neurologist Eric Lenneber had developed, before Chomsky, a universalist theory of language, arguing that all languages share the same underlying structure, that linguistic structures are largely innate, and that what is perceived as differences between specific languages is merely a superficial phenomenon that does not affect the cognitive process, which is universal in all human beings.

If we compare Chomsky's GU with Mentalese:
Some opinions and criticisms

Since its formulation by Fodor, the language-of-thought hypothesis (Mentalese) has provoked passionate debates among philosophers and cognitive scientists. The main objections are: Here are some views:
Propositional vs. imaginal

In cognitive psychology it is admitted that mental functions (including thinking) have a representational substrate. Two basic alternatives, in principle opposed, have been proposed: the propositional and the imaginal conception. In reality it is a dialectical confrontation, once again, of the two modes of consciousness: left hemisphere (LH) and right hemisphere (RH), between the external world (objective, rational and analytical) and the internal world (subjective, intuitive and synthetic). There is a third alternative, proposed by the connectionist school, and that is distributed representation: mental content cannot be localized because it is not stored as such, but is the result of the functioning of a network of connections.

The advantage of propositions is that they are very versatile. The disadvantage is that they require a large number of relations and inferences, many of them trivial. And there are mental phenomena such as qualia (the subjective qualities of conscious experiences, e.g., smelling a rose, feeling the color red, experiencing pain, etc.) that cannot be represented by the propositional resource. Propositional formalisms are intrinsically solipsistic, since they refer only to themselves.

Propositionalists postulate a Mentalese based on propositions. Mental states would be syntactic structures that encode meaning. Thought would consist of operations (computations) between syntactic representations, a process similar to the state transitions of a Turing machine.

The advantage of images is that they describe spatial relationships in a synthetic way. The disadvantage is that they do not allow formal or deductive computational processes. Because images belong to the subjective world and are more difficult to handle, attempts have been made to ignore them or at least to consider whether they are dispensable in the elaboration of a theory of cognition.

The topic of mental imagery has a long history in psychology. The first paradigm in psychology was of the mentalistic type and advocated mental imagery. Later, the possibility of the existence of thoughts without images was raised. With the arrival of the behaviorist school, the subject of mental images was absolutely avoided, as they were considered internal, subjective, unobservable and impossible to approach scientifically. Currently there is great interest in the subject of mental images, one of the questions raised being the nature of the images (schematic, structural, symbolic or abstract).

But the imaginal world is richer than the propositional one, since it is an inner, flexible and creative world. Mental imagery is a cognitive activity that allows us to perceive (internally) images of real objects that are not present. And the possibility of generating or creating images of imaginary entities (monsters, abstract figures, illogical or disproportionate relationships, etc.). As examples in the art world, we can cite Dali and Magritte. Imagination is more powerful than thought. Many scientific discoveries have come from imaginative intuitions. "Imagination is more important than knowledge. Knowledge is limited. Imagination encompasses the world" (Einstein).

Here are some views concerning the propositional-imaginal dialectic of mental representation:
Antecedents

Aristotle

Through Boethius we learn about Aristotelian theories of language and logic. Aristotle intuited the existence of a mental language (internal), related to the spoken language (external):
St. Augustine

St. Augustine also addressed the issue of the existence of a mental language, which he conceives as a unique language, common to all human beings:
Ockham

William of Ockham was the first philosopher in history to describe mental language at some level of detail, using grammatical and semantic categories: Ockham's mental language follows his principle of parsimony or conceptual economy (Ockham's famous razor), for it has a reductionist character. Mental terms are absolute and irreducible. All other categorematic mental terms (the connotative ones) can be reduced to absolute and syncategorematic terms.


MENTAL vs. Mentalese

Conclusions

With MENTAL the subject of the Mentalese hypothesis, the language of thought, is clarified:

Bibliography