Saussure's Second Course of Lectures on General Linguistics
from the notebooks of Albert Riedlinger and Patois
Trans: George Wolfe
- Second course of lectures 5 Nov. 1908 -- 24 June 1909 to eleven students.
- "Thus in a language there are many contradictory aspects.
There is nowhere where the language can be classified, there is no object
like it." (1a)
- "If therefore we consider the sphere in which the language lives,
there will always be the individual language and the social language.
Forms, grammars exist only socially, but changes start with an individual.
We cannot abandon one of the sides except by abstraction, and that always
carries a danger: that of attributing to a single side what belongs
just as much to the other, still within the same duality."(3a)
- Whitney, an american 1827-1894 who was an indianist/orientalist and
linguist; said language primarily is an institution.
- La Langue: "The Language is: a set of necessary conventions
adopted by the social body so as to permit the usage of the faculty
of language among individuals. The faculty of language is a fact
distinct from the language but which cannot be excersized without it.
By speech we designate the act of the individual putting his
faculty into practice by means of the social convention which is the
language. In speech there is an idea of the realization of what
is permitted by the social convention" (4a)
- again reiterates the spoken language alone is the object of linguistics.
- Semiology; a system of signs, prinicple characteristics:
signs, not symbols; symbols represent a connection between sign and
"All forms, all rites, all customs have a semiological character.
In the possible case in which the meaning of a custom has become completely
lost, we are in the same situation as that in which the words of a language
become unintelligible to speakers." (10)
"If you add a sign to the language you so far subtract from
the meaning of other signs" (12a)
therefore, the essential aspect:
- the arbitrary nature of the sign (no connection between sign and
- the purely negative and differential value of the sign
- The values of writing operate only as quantities opposed in a defined
system; the are oppositive, are only values through opposition.
There is a limit to the number of values.
- Complete indifference of the means of production of the sign.
"What is in the language escapes the individual or social will..."
"Language fundimentally has the character of a system founded
on oppositions, like a game of chess with the various combinations of
forces attributed to the different pieces, [and where the relative position
of the pieces, and not their intrinsic value is at stake] (19a)
"Seen from its internal side, the language thus strikes us as
not offering a concrete unit and yet we are not able to abandon the
notion that there is one, and it is the play of such units that makes
the language." (20a)
Phonic matter will always be in the same direction (one dimentional)
and does not admit the simultaneity of two signs. When we
speak of signs we immediately think of visual signs and we fall into
the misconception that the separation of signs is quite simple and does
not require an operation of the mind." (21a)
The material side of the sign is amorphous, which has no form in itself.
Image of a body of water and air above. As the airpressure changes,
it creates a wave. The wave is the unit, though nothing in itself,
but the interrelationship of water and air. Likewise, thought
and speech, thought the air, speech the water; both amorphous qualities,
together form language.
he's getting philosophical now: what is the nature of identity?
If I say messieurs and messieurs, what makes them identical; in fact
it is not, as the words are successive in time, seperate speech acts,
yet linguistically we wish to define them as identical. If we
repave a street, its the same street, or what about the 12:30 and 5
oclock express trains to Naples. Are they identical?
Where is the link linking messieurs said twice?
"There could be no identity if certain tacit conditions are not
agreed on in advance" (22a)
the two streets have an oppositive, or purely negative identity.
"The idea of unit would perhaps be clearer for some if we spoke
of meaningful units. But we must insist on the term 'unit': otherwise
we are liable to misconception and to believing that there are words
existing as units to which a meaning becomes attached. It is
on the contrary meaning which delimits words in thought." (24a)
In speech we here sound; thought defines words from sentences, and
sentenses from eachother. In themselves there is no logical break
between words or so forth.
- signs passively recieved from preceding generations
- signs will have as a characteristic to be transmitted in conditions
which have nothing to do with those which created it.
- The system of signs is materially altered in transmission, which
alters the relationship of sign to thought.
- The relationship of sign to thought is precisely what the sign is:
a double entity constituted by a succession of syllables to the extent
that a determinate meaning is attached to them; the sign is double.
Internal division of linguistic matters
- external study of linguistics: history and external description:
Anything which concerns language without entering into its system.
- internal study: touches on many "non-linguistic topics"
thus definition becomes negative.: what is capable of changing
- "The word is the most strongly delimited unit. The linguist
who wishes to delimit the word unit should try to figure out what this
separation is based on; which could provide the subject of a year's
course. There are writing systems which do not recognize word
separation. The illiterate (cook's letters) have no idea of the
exact separation of words. Such separation always comes back to
value and to identity through value; the unit is non-existent in advance,
outside of value." (30a)
- Two types of identities in language
- Diachronic: identity (moving) through time.
- Synchronic: those which constitute a state. [dhamma is
often translated 'state']. What belongs to a determinate instant
- idiosynchronic: for determinate instances of determinate languages
- diachronic identity may fall across several languages
- In phonetic change it is the sum of sounds which change [in this example
of oreos kata (coming from mountain down)] whereas here it is the unit,
the idea which has changed....
- Diachronic order = displacement of values / ...meaningful units
- Idiosynchronic order = determinate equilibrium of values as it is
established from moment to moment
- opposed like cinematic and static.
- "All phenomena are relationships among relationships. Or
let us talk rather of differences: everything is only a difference used
as an opposition and the opposition yields the value." (43a)
- It is only the sychronic which forms the system. The effect
of diachronic facts is to modify this system at every moment, but they
are not linked to each other, do not form a system among themselves,
are only a sum of isolated facts.
- "Only that thing is meaningful which has a synchronic difference
or synchronic facts to express it. It is difference which makes
something meaningful and it is meaning which also creates differences.
- "What is spatial must of course be translated by an idea of time..."
- "We need a sychronic fact in order to produce an analogy..."
- "For me there is no historical grammar; the terms clash: there
is no system which can straddle a succession of periods. What
is synchronically in one language is an equilibrium which is realized
from one moment to the next. By historical grammar what is meant
is diachronic linguistics, which is something else and is doomed never
to be grammatical." (62a)
- "In the language there are only differences and no positive qualities."
- "I have only wished to make the classification of everything
synchronic and I have divided it into a syntagmatic classification and
an associative classification..." (63a) "
- "...we are in two domains; one extends into a state of things,
is synchronic; the other extends in time."
- "The foundation of linguistics is dated from the first work of
F. Bopp, On the Sanskrit Conjugational System compared to that of
Latin, Greek, Persian, and Germanic, 1816." (72a)
- DumDUM DUUUUMMM...
- Father Coeurdoux (1767, Pondicherry) to Abbot Barthelemy, "How
comes it that in the Sanskrit language there is a large number of
words common with Greek and especially with Latin?"
- "The word Sanskrit: the samskrta language = ornate, ceremonial,
cultivated language, as opposed to the Prakrit language, the base,
natural idiom, is a dead language which has to be learned by study,
is in the same situation with respect to the popular dialects (Pali,
prakrit, etc.) as Latin with respect to the Romance Languages."
- W. Jones noted the similarity in 1786, calling Sanskrit in structure
"more perfect" than Greek or Latin.
- Bopp was the first to concieve of using one language to shed light
on another, to use one to analyse another.
- August Schleicher attempted to codify the science founded by Bopp.
- "We cannot say these [schleicher's] views satisfy us today; at
least we must not his tendency toward the general, the systematic.
A system, even if we have to abandon it later, is worth more than a
host of confused notions." (78a)
- Sanskrit in the mid 1800's was (erroneously) considered the SOURCE
of Latin-Greek-Celtic-Slavic (Indo-European) languages.
THe idea was that there was a primative indo-european original language
which was Sanskrit.
- "Simple reasoning indicates to everyone that all languages are
equally old...: since the dawn of time there has never been a language
which was not the continuation of what was spoken the day before..."
- "Aryas (old Vedic) opposes Indo-European race to every population
in India which was not Indo-European (an-aryas). Only the three
highest castes were designated Aryan (in the fourth, soudras. the blood
is not European)." (95a) [So is Buddhism racist? What
is the significance of the use of 'Ariya' in Pali??? What was
- more interesting still... Ariya ~ Eran ~ Iran [airyanam]
- We could say Aryan instead of Indo-European, but that designation
is better left to Indo-Iranian...
- [Early linguists adhered to the dubious notion of the primacy of geographic
transport of language] "We could designate this the Theory of Swarms.
As many languages as there are peoples, as many peoples as there are
swarms; these swarms separated from a primative center. These
swarms take us back to a primative hearth and so it was necessary to
resolve directly the question of point of departure. No one denied
that it was from Asia that this spreading out of peoples began (plateau
of Pamir!) There was also one of those mystical ideas, that migrations
had to go from east to west, that they advance toward the setting sun.
Why, no one knows." (97a)
- the fatal flaw here is that it doesn't take into consideration diversification
on the spot into dialects.
- Dutch (Hollandais), Dutch in the 10th century was a german dialect.
- "Hindu -- Aryans -- according to the Rig-Veda are still in Punjab.
The name of the Ganges occurs a single time in book 10, which is recent."
- Celts sacked the temple of Delphi, wandered about the Danube, into
Italy in the 5th cent BC. [crazy, man!]
- 3rd cent BC Germans were in the Black sea area, far from the Rhine...
they were still moving too...
Fin M. Riedlinger's notebooks on the second course...
Begin M. Patois, searching for difference:
- "Grammar is concerned with the functions of forms whereas morphology
determines the state of these forms. This distinction is basically
illusory. Units cannot be separated except by meaning, and vice
- "In a language state there is only a play of differences; but
it is always a question of differences which operate within a relative
unity (this is what coordinates them)." (ibid)
- Internal treasury (storehouse) = Associative units or groups in the
sense of families
- Discourse = Discursive units which are produced in discourse; groups
in the sense of syntagmas
- "Syntagma groups: The idea of a spatial limitation is immediately
evoked. THe condition of this order is extention and this is a
simple condition; language only has one dimention. There
is only one way to make a syntagma: by linear succession. (space, spacial
should be understood in time because we are dealing with spoken language).
- 'desireux' is a syntagma becuase we can distinguish desir-eux.
Similarly, 'que vous dit-il' is a syntagma for the same reason.
- syntagmatic relates to what is spoken, as opposed to associations,
which relate to what is thought.
- "This word Indo-European is not well chosen. We should
say Arya-European because 'aryas' among the Hindus is said of those
who speak Hindu related to our language" (161).