"Samizdat: Publishing the Forbidden."

Tag: universal grammar

Eco: Dee’s Magic Language


Florence Estienne Méric Casaubon (1599-1671), A True and Faithful Relation of what Passed for Many Yeers between Dr. John Dee [ . . . ] and Some Spirits, London, 1659. This work is in the public domain in its country of origin and other countries and areas where the copyright term is the author’s life plus 100 years or less. 

“In his Apologia compendiaria (1615) Fludd noted that the Rosicrucian brothers practiced that type of kabbalistic magic that enabled them to summon angels. This is reminiscent of the steganography of Trithemius. Yet it is no less reminiscent of the necromancy of John Dee, a man whom many authors considered the true inspirer of Rosicrucian spirituality.

In the course of one of the angelic colloquies recorded in A True and Faithful Relation of what Passed for Many Yeers between Dr. John Dee [ . . . ] and Some Spirits (1659: 92), Dee found himself in the presence of the Archangel Gabriel, who wished to reveal to him something about the nature of holy language.

When questioned, however, Gabriel simply repeated the information that the Hebrew of Adam, the language in which “every word signifieth the quiddity of the substance,” was also the primal language–a notion which, in the Renaissance, was hardly a revelation.

After this, in fact, the text continues, for page after page, to expatiate on the relations between the names of angels, numbers and secrets of the universe–to provide, in short, another example of the pseudo-Hebraic formulae which were the stock in trade of the Renaissance magus.

Yet it is perhaps significant that the 1659 Relation was published by Meric Casaubon, who was later accused of partially retrieving and editing Dee’s documents with the intention of discrediting him.

There is nothing, of course, surprising in the notion that a Renaissance magus invoked spirits; yet, in the case of John Dee, when he gave us an instance of cipher, or mystic language, he used other means.

In 1564, John Dee wrote the work upon which his contemporary fame rested–Monas hieroglyphica, where he speaks of a geometrical alphabet with no connection to Hebrew. It should be remembered that Dee, in his extraordinary library, had many of Lull’s manuscripts, and that many of his kabbalistic experiments with Hebrew characters in fact recall Lull’s use of letters in his art of combination (French 1972: 49ff).

Dee’s Monas is commonly considered a work of alchemy. Despite this, the network of alchemical references with which the book is filled seems rather intended to fulfill a larger purpose–that of explicating the cosmic implications deriving from Dee’s fundamental symbol, the Monad, based upon circles and straight lines, all generated from a single point.


John Dee (1527-1609), Monas hieroglyphica, 1564, held in the Bibliothèque nationale de France. The Monad is the symbol at the heart of the illustration labeled Figure 8.1 in Eco’s  The Search for the Perfect Language, Oxford, 1995, p. 186.

In this symbol (see figure 8.1), the main circle represented the sun that revolves around its central point, the earth, and in its upper part was intersected by a semi-circle representing the moon.

Both sun and moon were supported on an inverted cross which represented both the ternary principle–two straight lines which intersect plus their point of intersection–and the quaternary principle–the four right angles formed at the intersections of the two lines.

The sum of the ternary and quaternary principles constituted a further seven-fold principle, and Dee goes even on to squeeze an eight-fold principle from the diagram.

By adding the first four integers together, he also derives a ten-fold principle. By such a manipulatory vertigo Dee then derives the four composite elements (heat and cold, wet and dry) as well as other astrological revelations.

From here, through 24 theorems, Dee makes his image undergo a variety of rotations, decompositions, inversions and permutations, as if it were drawing anagrams from a series of Hebrew letters.

Sometimes he considers only the initial aspects of his figure, sometimes the final one, sometimes making numerological analyses, submitting his symbol to the kabbalistic techniques of notariqon, gematria, and temurah.

As a consequence, the Monas should permit–as happens with every numerological speculation–the revelation of the whole of the cosmic mysteries.

However, the Monad also generates alphabetic letters. Dee was emphatic about this in the letter of dedication with which he introduced his book. Here he asked all “grammarians” to recognize that his work “would explain the form of the letters, their position and place in the alphabetical order, and the relations between them, along with their numerological values, and many other things concerning the primary Alphabet of the three languages.”

This final reference to “the three languages” reminds us of Postel (whom Dee met personally) and of the Collège des Trois Langues at which Postel was professor. In fact, Postel, to prove that Hebrew was the primal language in his 1553 De originibus, had observed that every “demonstration of the world” comes from point, line and triangle, and that sounds themselves could be reduced to geometry.

In his De Foenicum literis, he further argued that the invention of the alphabet was almost contemporary with the spread of language (on this point see many later kabbalistic speculations over the origins of language, such as Thomas Bang, Caelum orientis, 1657: 10).

What Dee seems to have done is to take the geometrical argument to its logical conclusion. He announced in his dedicatory letter that “this alphabetic literature contains great mysteries,” continuing that “the first Mystic letters of Hebrews, Greeks, and Romans were formed by God and transmitted to mortals [ . . . ] so that all the signs used to represent them were produced by points, straight lines, and circumferences of circles arranged by an art most marvelous and wise.”

When he writes a eulogy of the geometrical properties of the Hebrew Yod, one is tempted to think of the Dantesque I; when he attempts to discover a generative matrix from which language could be derived, one thinks of the Lullian Ars.

Dee celebrates his procedure for generating letters as a “true Kabbalah [ . . . ] more divine than grammar itself.”

These points have been recently developed by Clulee (1988: 77-116), who argues that the Monas should be seen as presenting a system of writing, governed by strict rules, in which each character is associated with a thing.

In this sense, the language of Monas is superior to the kabbala, for the kabbala aims at the interpretation of things only as they are said (or written) in language, whereas the Monas aims directly at the interpretation of things as they are in themselves. Thanks to its universality, moreover, Dee can claim that his language invents or restores the language of Adam.

According to Clulee, Dee’s graphic analysis of the alphabet was suggested by the practice of Renaissance artists of designing alphabetical letters using the compass and set-square.

Thus Dee could have thought of a unique and simple device for generating both concepts and all the alphabets of the world.

Neither traditional grammarians nor kabbalists were able to explain the form of letters and their position within the alphabet; they were unable to discover the origins of signs and characters, and for this reason they were uncapable (sic) to retrieve that universal grammar that stood at the bases of Hebrew, Greek and Latin.

According to Clulee, what Dee seems to have discovered was an idea of language “as a vast, symbolic system through which meanings might be generated by the manipulation of symbols” (1988: 95).

Such an interpretation seems to be confirmed by an author absent from all the bibliographies (appearing, to the best of my knowledge, only in Leibniz’s Epistolica de historia etymologica dissertatio of 1717, which discusses him in some depth).

This author is Johannes Petrus Ericus, who, 1697, published his Anthropoglottogonia sive linguae humanae genesis, in which he tried to demonstrate that all languages, Hebrew included, were derived from Greek.

In 1686, however, he had also published a Principium philologicum in quo vocum, signorum et punctorum tum et literarum massime ac numerorum origo. Here he specifically cited Dee’s Monas Hieroglyphica to derive from that matrix the letters of all alphabets (still giving precedence to Greek) as well as all number systems.

Through a set of extremely complex procedures, Ericus broke down the first signs of the Zodiac to reconstruct them into Dee’s Monad; he assumed that Adam had named each animal by a name that reproduced the sounds that that each emitted; then he elaborated a rather credible phonological theory identifying classes of letters such as “per sibilatione per dentes,” “per tremulatione labrorum,” “per compressione labrorum,” “per contractione palati,” “per respiratione per nares.”

Ericus concluded that Adam used vowels for the names of the beasts of the fields, and mutes for the fish. This rather elementary phonetics also enabled Ericus to deduce the seven notes of the musical scale as well as the seven letters which designate them–these letters being the basic elements of the Monas.

Finally, he demonstrated how by rotating this figure, forming, as it were, visual anagrams, the letters of all other alphabets could be derived.

Thus the magic language of the Rosicrucians (if they existed, and if they were influenced by Dee) could have been a matrix able to generate–at least alphabetically–all languages, and, therefore, all the wisdom of the world.

Such a language would have been more than a universal grammar: it would have been a grammar without syntactic structures, or, as Demonet (1992: 404) suggests, a “grammar without words,” a silent communication, close to the language of angels, or similar to Kircher’s conception of hieroglyphs.

Thus, once again, this perfect language would be based upon a sort of communicative short-circuit, capable of revealing everything, but only if it remained initiatically secret.”

Umberto Eco, The Search for the Perfect Language, translated by James Fentress, Blackwell. Oxford, 1995, pp. 185-90.

Eco: The Pre-Hebraic Language


Athanasius Kircher (1602-80), Origins of the Chinese Characters, China Illustrata, or China monumentis, qua sacris qua profanis, nec non variis naturae & artis spectaculis, aliarumque rerum memorabilium argumentis illustrata (1667), p. 229. Courtesy of Stanford University. This work is in the public domain in its country of origin and other countries and areas where the copyright term is the author’s life plus 100 years or less. 

“Alongside these philosophical discussions, other inspired glottogonists (for whom the defeat of the Hebraic hypothesis was a consummated fact) were breaking new theoretical ground.

The explorers and missionaries of the sixteenth and seventeenth centuries had discovered civilizations, older than the Hebrews, which had their own cultural and linguistic traditions.

In 1699, John Webb (An Historical Essay endeavoring the Probability that the Language of the Empire of China is the Primitive Language) advanced the idea that, after the Flood, Noah had landed his Ark and gone to live in China.

Consequently, it was the Chinese language which held primacy. Furthermore, since the Chinese had not participated in the construction of the Tower of Babel, their language had remained immune from the effects of the confusio; Chinese had survived intact for centuries, protected from foreign invasion. Chinese thus conserved the original linguistic patrimony.

Ours is a story that proceeds through many strange anachronisms. Near the end of the eighteenth century, just at the moment when, quite unconnected with any form of the monogenetic hypothesis, a comparative methodology was about to emerge, there appeared the most gigantic attempt to date to rediscover the primitive language.

In 1765, Charles de Brosses wrote a Traité de la formation méchanique des langues. The treatise propounded a theory of language that was both naturalistic (the articulation of terms reflects the nature of things–sweet sounds designate sweet objects) and materialistic (language is reduced to physical operations, supernatural entities are seen as the result of linguistic play: cf. Droixhe 1978).

As part of this theory, however, de Brosses could not resist indulging in a series of speculations about the nature of the primitive language, “organic, physical, and necessary, that not one of the world’s peoples either knows or practices in its simplicity, but which, none the less, was spoken by all men, and constitutes the basis of language in every land” (“Discours préliminaire,” xiv-xv).

“The linguist must analyze the mechanisms of different languages, discovering which of those features arise through natural necessity. From this he may, moving through a chain of natural inferences, work his way back from each of the known languages to the original, unknown matrix.

It is only a matter of locating a small set of primitive roots that might yield a universal nomenclature for all languages, European and oriental.

Radically Cratylian and mimologist as it was (cf. Ginette 1976: 85-118), the comparative approach of de Brosses took the vowels to constitute the raw material in a continuum of sound upon which the consonants acted to sculpt out the intonations and the caesurae.

Their effect, often more visible to the eye than to the ear (remember the persistent failure to distinguish between sounds and letters), is to render consonantal identity the key criterion of comparative analysis.

Like Vico, de Brosses considered that the invention of articulated sounds had proceeded in step with the invention of writing. Fano (1962: 231; English tr., p. 147) sums up his theory very well:

“De Brosses imagines this process as follows: like the good school teacher who takes chalk in hand to make his lesson clearer from a didactic viewpoint, the cave man intermingled his discourses with little explicative figures.

If, for example, he wanted to say “a raven flew away and rested on the top of a tree,” he would first imitate the croaking of the bird, then he would express the flight with a “frrr! frrr!” and eventually take a piece of coal and draw a tree with a raven on top.”

Another Herculean effort in the cause of mimological hypothesis was that of Antoine Court de Gébelin, who, between 1773 and 1782, published nine quarto volumes, totaling over five thousand pages, giving to this opus–multiple, creaking, though not utterly devoid of interest–the title Le monde primitif analysé et comparé avec le monde moderne (cf. Genette 1976: 119-48).

Court de Gébelin knew the results of previous comparativist research. He also knew that the human linguistic faculty was exercised through a specific phonatory apparatus; and he was acquainted with its anatomy and physiology.

He followed, moreover, the doctrines of the Physiocrats, and when he sought to explain the origin of language, he did so through a re-reading of ancient myths, interpreting them as allegories describing the relation of man the farmer to the land (vol. I).

Writing, too, was susceptible to this sort of explanation. Although it was born before the separation of peoples, writing could be interpreted as having evolved in the time of the agrarian states, which needed to develop an instrument that would keep track of landed property and foster commerce and law (vol. III, p. xi) . . .

Yet there still shines Court de Gébelin‘s dream of uncovering the original language of the primitive world, the language which served as the origin and basis of a universal grammar through which all existing languages might be explained.

In the preliminary discourse to volume III, dedicated to the natural history of speech or the origins of language, Court de Gébelin affirmed that words were not born by chance: “each word has its own rationale deriving from Nature” (p. ix). He developed a strongly mimological theory of language accompanied by an ideographic theory of writing, according to which the alphabet itself is nothing but the primitive hieroglyphic script reduced to a small set of radical characters or “keys” (III, xii).

As a faculty based upon a determined anatomical structure, language might certainly be considered as God’s gift, but the elaboration of a primitive tongue was a human endeavor. It followed that when God spoke first to human beings, he had to use a language that they could understand, because it was a product of their own (III, 69).

To uncover this primitive language, Court de Gébelin undertook an impressive etymological analysis of Greek, Latin and French. Nor did he neglect coats of arms, coins, games, the voyages of the Phoenicians around the world, American Indian languages, medallions, and civil and religious history as manifested in calendars and almanacs.

As a basis for this original language he set out to reconstruct a universal grammar, founded on necessary principles, valid in all times and in all places, so that the moment that one of these principles was discovered lying immanent in any one language it could be projected into all the others.

Court de Gébelin seems, in the end, to have wanted too much. He wanted a universal grammar; he wanted the mother tongue; he wanted the biological and social origins of language.

He ended up, as Yaguello observes (1984: 19), by muddling them all together in a confused mass. To top it all off, he fell victim in the end to the siren call of the Celto-nationalist hypothesis which I shall be describing in the next section.

Celtic (being similar to oriental languages from which it originated) was the tongue of Europe’s first inhabitants. From Celtic had derived Greek, Latin, Etruscan, Thracian, German, the Cantabrian of the ancient Spaniards, and the Runic of the Norsemen (vol. V).”

Umberto Eco, The Search for the Perfect Language, translated by James Fentress, Blackwell. Oxford, 1995, pp. 91-5.

Eco: Dante and Universal Grammar


Cornelisz Anthonisz, The Fall of the Tower of Babel, 1547. The text at top right reads “Bablon / Genesis 11.” The text in the top left banner reads, “When it was at its highest / it should not do fall.” The stone at the bottom left reads “1547.” This work is in the public domain in its country of origin and other countries and areas where the copyright term is the author’s life plus 100 years or less.  

“One solution to the problem has been proposed by Maria Corti (1981: 46ff). It is, by now, generally accepted that we cannot regard Dante as simply an orthodox follower of the thought of St. Thomas Aquinas.

According to circumstances, Dante used a variety of philosophical and theological sources; it is furthermore well established that he was influenced by various strands of the so-called radical Aristotelianism whose major representative was Siger of Brabant.

Another important figure in radical Aristotelianism was Boethius of Dacia, who, like Siger, suffered the condemnation of the Bishop of Paris in 1277. Boethius was a member of a group of grammarians called Modistae, and the author of a treatise, De modis significandi, which–according to Corti–influenced Dante, because Bologna was the focal point from which, either through a stay in the city, or through Florentine or Bolognese friends, such influences reached Dante.

The Modist grammarians asserted the existence of linguistic universals–that is, of rules underlying the formation of any natural language. This may help clarify precisely what Dante meant by forma locutionis. In his De modis, Boethius of Dacia observed that it was possible to extract from all existing languages the rules of a universal grammar, distinct from either Greek or Latin grammar (Quaestio 6).

The “speculative grammar” of the Modistae asserted a relation of specular correspondence between language, thought and the nature of things. For them, it was a given that the modi intelligendi and, consequently, the modi significandi reflected the modi essendi of things themselves.

What God gave Adam, therefore, was neither just the faculty of language nor yet a natural language; what he gave was, in fact, a set of principles for a universal grammar. These principles acted as the formal cause of language: “the general structuring principle of language, as regards either the lexicon, or the morphological and syntactical components of the language that Adam would gradually forge by living and giving names.” (Corti 1981: 47).

Maria Corti’s thesis has been vehemently contested (cf., in particular, Pagani 1982; Maierù 1983). It has been objected that there is no clear proof that Dante even knew the work of Boethius of Dacia, that many of the analogies that Maria Corti tries to establish between Dante’s text and Boethius cannot be sustained, and that, finally, many of the linguistic notions that one finds in Dante were already circulating in the works of philosophers even before the thirteenth century.

Now, even if the first two objections are conceded, there still remains the third. That there were widespread discussions of the subject of universal grammar in medieval culture is something that no one, and certainly not Corti’s critics, wishes to place in doubt.

As Maierù puts it, it was not necessary to read Boethius to know that grammar has one and the same substance in all languages, even if there are variations on the surface, for this assertion is already found in Roger Bacon.

Yet this, if anything, constitutes proof that it was possible that Dante could have been thinking about universal grammar when he wrote DVE. If this is so, he could have conceived of the forma locutionis given by God as a sort of innate mechanism, in the same terms as Chomsky’s generative grammar, which, interestingly enough, was inspired by the rationalist ideas of Descartes and sixteenth-century grammarians who, in their turn, had rediscovered the ideas of the medieval Modistae.

Yet if this is all there is to it, what is the point of the story of Babel? It seems most likely that Dante believed that, at Babel, there had disappeared the perfect forma locutionis whose principles permitted the creation of languages capable of reflecting the true essence of things; languages, in other words, in which the modi essendi of things were identical with the modi significandi.

The Hebrew of Eden was the perfect and unrepeatable example of such a language. What was left after Babel? All that remained were shattered, imperfect formae locutionis, imperfect as the various vulgar Italian dialects whose defects and whose incapacity to express grand and profound thoughts Dante pitilessly analyzed.”

Umberto Eco, The Search for the Perfect Language, translated by James Fentress, Blackwell. Oxford, 1995, pp. 43-5.

Eco: Search for the Perfect Language, 2.

Pieter Brueghel the Elder (1526/1530-1569 CE), The Tower of Babel. Brueghel painted three versions of the Tower of Babel. One is kept in the Museum Bojimans Van Beuningen in Rotterdam, the second, this one, is held in the Kunsthistoriches Museum in Vienna, while the disposition of the third version, a miniature on ivory, is unknown. Faithful reproductions of two-dimensional public domain works of art are public domain.

Pieter Brueghel the Elder (1526/1530-1569 CE), The Tower of Babel. Brueghel painted three versions of the Tower of Babel. One is kept in the Museum Bojimans Van Beuningen in Rotterdam, the second, this one, is held in the Kunsthistoriches Museum in Vienna, while the disposition of the third version, a miniature on ivory, is unknown. Faithful reproductions of two-dimensional public domain works of art are public domain.

“Beyond this, I have decided to consider only projects concerning true and proper languages. This means that, with a bitter sigh of relief, I have decided to consider only the following:

  1. the rediscovery of languages postulated as original or as mystically perfect — such as Hebrew, Egyptian or Chinese;
  2. the reconstruction of languages postulated, either fancifully or not, as original or mother tongues, including the laboratory model of Indo-European;
  3. languages constructed artificially for one of three ends: (a) perfection in terms of either function or structure, such as the a priori philosophical languages of the seventeenth and eighteenth centuries, which were designed to express ideas perfectly and to discover thereby new connections between the diverse aspects of reality; (b) perfection in terms of universality, such as the a posteriori international languages of the nineteenth century; (c) perfection in terms of practicality, if only presumed, such as the so-called polygraphies;
  4. more or less magic languages, whether they be discovered or fabricated, whose perfection is extolled on account of either their mystic affability or their initiative secrecy.

By contrast, I can give no more than bare notice to any of the following:

  1. oneiric languages, not expressly invented, such as the languages of the insane, or of trance states, or of mystic revelations (like the Unknown Language of Saint Hildegarde of Bingen), as well as all the cases of glossolalia or xenoglossia (cf. Samara 1972); Goodman 1972);
  2. fictitious languages, either in narrative (from Rabelais to Foigny up to Orwell’s ‘Newspeak’ and Tolkien), or in poetry (like Chlebnikov’s transmental speech). In the majority of these cases, we are presented with only short stretches of speech, supposedly representing an actual language, for which, however, there is provided neither a lexicon nor a syntax (cf. Pons 1930, 1931, 1932, 1979; Yaguello 1984).
  3. bricolage languages, that is languages that are created spontaneously by the encounter of two linguistically distinct cultures. Typical examples are the pidgins arising in areas of colonialism. As cross-national as they may be, they are not universal. They are, rather, partial and imperfect because they have a limited lexicon and an oversimplified syntax; they are used to facilitate simple activities such as barter, but are unable to express higher types of experience (cf. Waldman 1977);
  4. natural tongues or jargons serving as vehicular languages in multilingual zones. An example of such a language of exchange might be Swahili, the lingua franca of large areas of East Africa. Modern English would be another example. French was formerly an example, if one considers that, during the Convention, the Abbé Gregoire revealed that, out of a population of twenty-six million, fifteen million French men and women spoke a language other than that of Paris (Calvet 1981: 110);
  5. formal languages whose use is limited to special scientific purposes, such as the languages of chemistry, algebra and logic (these will be considered only as they derive from projects defined by category 3(a) above;
  6. the immense and delectable category of the so-called fous du language (see, for example, Blavier 1982; Yaguello 1984). Admittedly, in such cases it is not always easy to distinguish between technical insanity and mild glottomania, and many of my own characters may sometimes show some aspects of lunacy. Still, it is possible to make a distinction. We will not consider belated glottomaniacs. Nevertheless, I have not always been able to keep down my taste for whimsicality, especially when (even though the belatedness was hardly justifiable) those attempts had, anyway, a certain, traceable, historic influence, or, at least, they documented the longevity of a dream.

Similarly, I do not claim here to examine the whole of the researches on a universal grammar (except in cases in which they clearly intersect with my topic), because they deserve a separate chapter of the history of linguistics.

Likewise, this is not (except, again, where the subject intersects with that of the perfect language) a book about the secular, or rather, millennial, question of the origins of language.

There are infinite discussions on the origins of human language which do not consider the possibility or the opportunity of returning back to the language of our origins, either because they assume that it had definitely disappeared, or because they consider it as radically imperfect.

Finally, were it up to me to decide under which heading this book should be filed in a library catalogue (an issue which, for Leibniz, was bound up with the problem of a perfect language), I would pick neither ‘linguistics’ nor ‘semiotics’ (even though the book employs semiotics as its instrument, and demands a certain degree of semiotic interest from its reader).

I would rather pick ‘history of ideas.’ This explains why I make no attempt to construct a rigorous semiotic typology for the various types of a priori and a posteriori languages: this would require a detailed examination of each and every project, a job for students of what is now called ‘general inter linguistics.’

The present book aims instead at delineating, with large brushstrokes and selected examples, the principle episodes of the story of a dream that has run now for almost two thousand years.”

Umberto Eco, The Search for the Perfect Language, translated by James Fentress, Blackwell. Oxford, 1995, pp. 2-5.

%d bloggers like this: