American Journal of Computational Linguistics 
THE YoERKISH LANGUAG~E 
Department of Psychology 
University of Georgia 
and 
Yerkes ~egional Primate Research Center 
Microfiche $12 
Copyright 1975 
Association for Computational Linguistics 
Abstract 
Yerkish, the language descritr4d in this paper, was devigrred for 
the purpose of exploring the extent to which non-human organisms (e.g. 
great apes) could be t-,rougiit to acquire linguistic skills. First 
attempts at teaching a spoken language to non-human primates had failed, 
app3rently because ol tlle animals' incapacity vocally. to produce tlle 
phonemes of a natural language. Subsequent work (Gardner G Cardner, 1971; 
I'remack, 1971) demonstrated that colr.munication could be achieved Ey means 
of visual signs or symbols. 
Yerkish is a visual language with a lexicon of gsaplric word symbols 
(lexigrams), each of whicl;l is a combination of discrete recursive-design 
elements. Each lexigram is represented on one of 125 keys of a keyboard, 
Sentences are formed by pressing keys in successive order. Sentence 
length, at present, is limited to seven lexigrams. Input from the 
keyboard is monitored and recorded by a computer that, contains in its 
core the lexicon, A parser, and certain response capabilities. 
The grammar is strictly interpretive and was derived fror t1:e 
1 
correlational' systkm implenlented in the ?Iult is tore parser for 1:n~lisP. 
sentences (von Clasersfeld.& Pisani, 197-0). ?he parser works on the 
basis of essentially non-lingui~tic classifications of items arkd 
relational concepts (tables of the operational classes are provided in 
this Faper). It produces a structural analysis in terms of imr?.ediat,e 
constituents. If an input string yields one comprehensive structure, 
the string is deemed grammat.ica1. The first- lexigram functions as a 
sentence marker indicating the mood of the utterance: affirmative, 
negative, interrogative, and imperative. In the case of certain 
requests (imperatives), the conputer activates mechanical devices that 
fulfil the request (dispensing food, drink, toys, playing music or a 
movie, etc.) Given this response capability, a 24-hour learning situation 
is created in which there .is always some 'potential incentive for the 
animal to use linguistic communication. 
Yerkish lexigrams and sentences are easily translated into English, 
but semantics and sentence structure are somewhat restricted. Lexigrarns 
alvays have only one meaning of the corresponding English and 
the rules for their concatenation \$ere designed to reduce ambiguities to 
a minimum. 
The paper explains deviations from ~nglish grammar by means of 
structural diagrams and demonstrates that, in spite of the many simpli 
f ications, Yerkish allows for embedded clau-ses and is, indeed, capable 
of expansion in many directions. 
? 
Three examples of 
conversqtions' with the young female chimpanzee 
that is being taught the use of Yerkish are added as an appendix to 
the paper. 
TABLE OF CONTENTS 
Ustract .......................... 2 
Introduction ........................ 5 
Background ......................... 8 
The Communication Facility ................. 11 
The Yerkish Lexigrams ................... 14 
.......... Interpretive versus Descriptive Grammar 16 
.............. A Restrictxd Universe of Discourse 18 
Technical Constraints ................... 19 
................... The Grammar of Yerkish 22 
.............. Peculiarities of Yerkish Grammar 31 
Conclusion ......................... 38 
References ......................... 40 
TABLES 
1 . Design Elements .................... 42 
2 . Semantic Color-Coding of Lexigrams ........... 43 
3 . Operational Lexigram Classes .............. 44 
4 . Operational Correlators ................ 47 
5 . Assignation of Correlator Indices to L-igram Cldsses . 51 
6 . Assignation of Correlator Indices to Products 
(Reclassification) ................ 52 
APPENDIX 
................. 'Conversations' with Lana 53 
The Yerkish Language for Xon-Human Primate's 
- - 
(1) 
Introduction 
There are .several reasons why it would be cumbersome and even some- 
what misleading to organize a description of the language used in our 
communication study with non-human primates, according to the linguist's 
traditional pattern, in three nore or less idependent sections dealing, 
respectively, with granunar, lexicon,. and semantics. Yerkish is an 
artificial language that was designed for a specific and peculiar purpose 
to explorbe to what extent apparentry non-linguistic organisms could 
acquire linguistic skills if they were placed in an environment in 
which the use of linguistic communication would be to their advantage. 
Being an artificial language, the design of Yerkish was subject to 
constraints that are rather different from those that may or may not 
have impinged upon the development of natural languages. Conputational 
Research supported by National Ins ti tu tes of Health grants ID-0616 
and RR-00165 and., carried out at the Yerkes Regional Primate Research 
Center, Emory University (Atlanta), by a team of seven specialists 
from three actively participating institutions: Duane F:. Rumbaugk, 
PC 
initiator and Principal Investigator of the project, Timothy - V. - dill 
and Josephine Brown (Dept . of Psychology, Georgia State Cniversity) 
are responsible for behavioral design and experimentation. Harold 
Warner and Charles - Bell (Yef1:es Regional Primate Research Center) 
are responsible for design and engineering of the interface with 
the computer and the eledtromechanical devices. Ernst von Glasers- 
feld and Pier Paolo - Pisani (Dept. of Psychology and. Computer Center, 
University of Georgia) are responsible for the design and computeri- 
sation of the Yerkish language, its grammar, and the 'fl~l~istore 
linguists will, I am sure, agree that natural languages would have 
different grammars and different interpretive rules if, from the very 
beginning of their evolution, they had had to be intelligible to a 
computer. And that is what Yerkish had to be. For reasons that will 
become clear when we discuss the research background of the project, 
the introduction of a computer as m~nitor of the communication system 
was one of the salient feature of this research effort (Rumbaugh -- et al., 
1973a). 
Other constraints in the development of the Yerkish language will be 
discussed at those points in the exposition where their explication 
seemed most appropriate. I have tried to concentrate specific aspects 
under indicative subtitles. I have no illusions that this has been 
wholly successful. lly rmin goal, however, was to give the reader as 
complete as possihle a picture, not only of what was done, but also of 
why it was done. The instrumental aspect of Yerkish as the linguistic 
vehicle in an experimental communication study must be kept in mind at 
all times; much of what follows in these pages can make sense only if it 
is put into that perspective. 
Another point that I should like to stress is that. the artificial 
language on which this paper reports is only one of several major 
efforts that made this communication study possible. Such success as we 
have had is the result of. team work in the fullest sense of that term. 
The project would never have got off the ground if it had not been for 
the -continuous patient collaboration 6f seven rather heterogeqeous specia- 
lists from three different institutions (see footnote l) and, last but 
not least, for the perseverance of our f.emale chimpanzee Lana. (Fig. 1: 
Figure 1 
The chimpanzee Lana worl'ing at her lceyboard 
Background 
Yerkish is a visual language of graphic word-sy~tols, designed for 
research in .communication with non-human primates and, possi.bly, as a 
substitute vehicle for humans who, for physiological reasons,-could not 
acquire a spoken language. Lieberman(l968) , on the basis of anatomical 
investigations, .came to the conclusion that the vocal apparatus of the 
great apes precludes the production and modulation of many of the 
phonemes that make up the repertoire of human languages. Forty years 
earlier, Robert Yerk-es (1925), the founder of the first primatological 
research institute, in whose honor we have named our language, had 
already observed this vocal handicap of the non-human primates. In the 
intervening years, the Zailure of several long-term efforts to teach a 
chimpanzee English, Japanese, or Russian, empirically confirmed his 
observation '(for a review, cf. Floog and MelnechuE., 1971). 
The fact, however, that the great apeas are barred from speak.-ing 
a language does not necessarily mean that they could not understand one, 
nor that they could not learn to use a 1in.guistic co~munication systent 
that functions in another sensory modality. There still are, of course, 
It 
scholars who, defining language" from a rather anthropocentric point of 
view, refuse to allow the term for any cornrr.unication system that does not 
use the vocal-auditory mode of transmission. Among otl~er tl~ings , this 
would mean that programming languages and other silent communication 
systems could never be called "language", no matter lla~ adequately they 
might be described in terms of lexicon, syntax, and semantics. Today, 
It 
there seems to be a gro~~~ing consensus that this restriction of language" 
to acoustic systems is not a stientifically necessary or useful one 
(Ploog and Melnechucli, 1971 : 640; Lyons, 1972 : 64) . Interest, thus, 
has shifted from the question whether or not other organisms can learn 
to speak a language, to the question whether or not they can learn to 
handle a cornmunicati~n system that is linguistic in its structure. 
Given that there seems to be no compelling evidence that any non- 
human species on this earth has, in fact, developed a c~~unication 
11 
system that could legitimately be called language", one might ke 
inclined to thinlc that attemts t'o teach a Language to a non-human 
organism are necessarf-fg ci-owmd Bail. This conclusion, however, 
would be quite unwarranted. Animal trainers in circuses and in the 
laboratory have shown beyond all doubt that many species have a potential 
for the acquisition of skills t?hich no one, wl~o observed the species 
in the wild, would suspect. The fact is that the behaviors an 
organism manifests in a given environment constitute under all circum- 
stances only a subset of the behaviors which the organism could acquire 
in different environments (Lorenz, 1974). In the area of cognitive 
skills, for instance, ~ohler's investigations (1925) already indicated 
that the great apes had been drastically underrated. Since then, and 
up to the demonstration of "higher mental functions" by Viki (Hayes 
and Nissen, 1956/1971) , especially the cl~impanzee's intellectual 
reputation has continuously grodn (P.umbaug11, 1970) . Thus it has, indeed, 
become more and more pertinent to ask just how far a chimpanzee (or other 
great ape) could be brought in the acquisition of linguistic skills 
which do - not require vocal speecl~. 
T.he success of the Gardners (1969, 1971) with +-heir chimpanzee \\'ashoe 
is so well known that there is no need to rei,terate the description of 
their pioneering work. Using American Sign Language (ASL) as a vehicle, 
they established irrefQtably that an infant chimpanzee can be taught to 
communicate very effectively; and there would seem to be no reason why, 
given a conducive environment, Washoe's communicatory skill should not 
continue to grow as she develops towards intellectual maturity. 
It has been repeatedly objected, however, that Washoe's successful 
communications are as yet no proof that she has acquired "language" 
(e.g. Lenneherg, 1971; Brown, 1971). Piost of the skepticism about 
washoe's linguistilc accomplishments is based on the argument that the 
strings of signs she produced do not manifest syntactic competence, 
When Washoe was introduced to ASL, no rigid rules of sign-order were 
observed and the relational semantics (which, for instance in English, 
is taken care of to a large extent by word-order) was left implicit in 
the communicatory event and had to be intuitively gleaned from the 
situational context by the observer (Gardner and Gardner, 1971). Since 
a language user's compliance with the syntactic rules of the language 
an important criterion in the evaluation of his linguistic performance, 
the apparent tack of such rules in ASL made it - a priori questionable 
whether Washoe's or, indeed, any other ASL-user's stringing together of 
I I 
signs could be considered syntactic and thus evidence of language" on 
the theoretical level. In addition to that, a l'ack of syntactic rules 
is the- very reason why Washoe's ,~~mmunications could not contain many 
relational indications. For instance, since the sign system taught to 
Washoe had no consistent means for designating actor and patient in 
activity situations (comparable to, say, the subject'-yerb-ob ject sequence 
in many natural languages), the assignation of these roles was left 
to the intuition or the common sense of the receiver. 
In retrospett it is easy to see t-hat this relative lack of syntactic 
rigidity would supply critics with arguments that.tend to diminish 
claims with regard to washoe's linguistic competence. 
On the other 
hand, it is equally clear that the GardnersS when they started on 
their splendid enterprise., were concerned abov'e all with the formidable 
task of establishing a viable form of communication with a chimpanzee, 
and they could not possibly have foreseen all the theoretical reasons 
why linguists and philosophers of language might doubt that the 
communication system they chose, and the \\ray. VJashoe was going to use 
it, should be called "linguistic". Hence I should like to emphasize 
that my attempt to clarify the syntactic problem is in no way intended 
as a criticism of the work accomplished with Washoe, but solely to 
throw some light on the several ways in which our project, staxted a 
few years' l'ater, was able to benefit from the ~ardners' effort. 
The Cornmunitation Facility 
-, 
The basic idea of our project at the Yerkes Primate Research Center 
was the introduction of a computer as a thoroughly objective monztor of 
all linguistic transactions. This solved sev-era1 problems at once. In 
the first -place, it eliminated the proble~ of subjective or intuitive 
evaluation of the grammatical correctness of the experirental animal's 
linguistic products. 
Incorporating a reduced and suitably adapted 
Multistore Parser (von Glasersfeld and Pisani, 1970), the computer can 
t I 
objectively" judge grammaticality. 
An input string either conforms to 
legitimate syntactic structure, dr it does not. There cannot be any 
doubt either way. Second, the computer has no difficulmty in recording 
every input and transaction that takes place, be it grammatical or not. 
Third, thanks to the computer, the communication facility can be kept 
in operation twenty-four hours a day, without the forbidding cost of 
several shifts of techinicians and- observers. 
In order to turn the communication facility into a learning 
environment that could af least to some extent operate without the 
presence of a.human a system of automatic responses was implemented. 
By activating one of a set of machine-commanded dispensers, the computer 
can satisfy a number of requests, provided these requests are correctly 
formulated by the experimental animal. So far, the automatic responses 
are limited to the dispensing of various foods and drinks, to openivg 
and shutting 8 window., activating a movie and a slide projector as 
well as a tape player. In .the future we hope to add something of a 
question-answering system in order to enable the computer To respond 
verbally to some questions and, perhaps, also to give some feedback with 
regard to errore made in the subject's linguistic input. 
A full description of the communication facility, as it is in oper- 
ati,on at present, has been published elsewhere (Rumbaugh -- et al., 1973a), 
Here we shall be mainly concerned with the Yerkish language. A quick 
survey of the main components of the installation will have to suffice. 
Input to the system is effected by means of a keyboard of maximally 
125 keys, arranged in vertical panels of 25 each. Four such panels are 
in use at present, corresponding to a total of 100 E'eys. Each key 
represents one lexigram, i.e., a geometrical design which constitutes 
a word-symbol (lexical item) of the ~erkish language. Depression of 
a key activates the correspoqding item in the computerized lexicon 
which is permanently incorporated in the llultistore parser. The spatial 
arrangement of the lexigrams in the keyboard can be easily reshuffled 
(to prevent the experimental animal from acquiring a fixed motor pattern). 
To switch on the system, a horizontal bar, mounted \dell above the 
keyboard, has to be pulled down. The bqr has to be held down continuously 
thfoughout the input of a message. Lana, the female infant Fhimpanzee 
with whom we have bees working, does this by hanging on to the bar with 
one hand while using the other to press keys. If the system is switched 
on and several keys are then pressed in successi.on, ending with the 
I1 period" key (bl~e "end-of-message" signal for the computer) , the parser 
11 
takes this sfring as a- sentence" and analyzes it in order to establish 
whether or not is is gramnatikally correct. If the input string is a 
grammatically correct request, the machine also determines the object 
of the request and, if it is within the range of automated responses, 
satisfies the request by activating the r'elevant dispenser ar mechanism. 
Regardless of the outcore of the grmatical analysis, the machine 
prints out the English=word tor-responding to each lexigram that has 
been activated and records, at the end of the string, whether or not it 
was found to be correct. 
Directly above the keykoard in the experimental chamber, there is a 
row of seven! small projectors in which the geometric designs of the 
lexigrams appear, one by one from left to right, as their keys are 
being pressed on the keyboard. 
This provides Lana with feedback as to 
the part of the message that has already been typed in, and also with a 
linear representation of the string she is composing. P signallight, 
I1 
on the right of the projectors, lights up when the period" key has been 
pressed and terminates the message. 
Above this first row of projectors there is a second similar one 
which serves to display messages- that are seqt. in ta Lana from a second 
keyboard in the technicianfs station outside &anafs chamber. ljessages 
originating fromthe4technicianfs keyboard are also recorded the 
I I 
computer, but they are marked by a code symbol as operator's messages" 
and cannot be confounded with ~ana's linguistic production. 
Thue Yerkish Lexigrams 
The original constraints under which the Yerkish language was to 
be designed were eskentially three. 
1) Drawing on the experience of the Cardners (1969, 1971) and Premack 
(1971), Yerkish had to be a visual language with a lexicon of 
unitary word-symbols that could be represented singly on the keys 
of a keyboard. 
2) Both lexical items and sentence structure were to be as univocal as 
possible, because this, on the one hand, would facilitate the 
automatic parsing of input and, on the other, it was expected to 
make acquisition of the lariguage easier for our subjects. 
3) The structure of Yerkish was to be close enough to English 'to 
allow word-by-word translation, in order to make participation in 
communication- events, as well as their evaluation, maximally ac- 
cessiblk to technicians and observers. 
For a few weelcs at the very-outset of the enterprise, the author 
revelled in dreams of an ideal language in which each word was. to be 
composed of semantically significant pleremes (Hockett, 1961).   here 
were to be individual design elements designating the more important 
recurrent semantic categories, and each concept available in the ~erkish 
universe of discourse was to be represented by a lexigram (i,e. the 
visualfgraphic caunterparts to words in spoken languages) composed of 
design elements &ich, in their own right, would designate the major 
semantic categories to which the concept belonged. Thus, for instance, 
as the American Indian language Yuehi (Crawford, 1973) has a morpheme that 
recurs. in any word that designates a part of the human body, every Yerkish 
lexigram designating a part of the primate body would have contained 
a specific design element. Given that the Yerki,sh lexicon was, in any 
case,to co.ntain no more than two or three hundred lexigrams, it 
seemed feasible to cover at least the major semantic categories with a 
hundred or so design elements. The reason for doing this was, of course, 
that such a language would have been an invaluable instrument for testing 
our subject's c'lassificatory skill and processes of concept formation. 
The dream was soon shattered by fiechn.ica1 res tr.ictions. The feedback 
projectors above the keyboard had to be such that each one of them could 
display every lexigsam of the language. Within our budget, this could be 
achieved only if all lexigrams were designed in such a way that they could 
be generated by combining design elements of,a common set limited to 
twelve. 
Vnder these circumstances it was obviously impossible to main- 
tain the individual design elements semantically constant and a drastic 
compromise had to be made. By choosing nine graphic elements that could 
be readi-Ly superimpos-ed, one over the other, and thre6 basic colours, a 
little additional flexibility was gained (see Table 1.). By "mixing1' 'the 
three basic colours we could generate seven discriminable hues. Together 
with black (absence of colour) , this gave us eight background features, 
and these could be used to colour-code at least some important conceptual 
categories (see Table 2). 
Interpretive versus Descriptive Grammar 
I I 
The grammar of Yarkdsh is a direct derivative of the correlationa.1" 
'grammar that was implemented some years ago in the blultistore parser for 
English sentences (von Gldsersfeld, 1964, 1965, 1970; von Glasersfeld 
and Pisani, 1968, 1970). It is, therefore, strictly an interpretive 
I I I I 
grammar and lays no claim to ge.nerativeU properties, nor is it trans- 
formational" in the Chomskyan sense of that term. 
Tn the hope that it might dispel some misunderstandings that have 
haunted the development of correla.tiona1 gr.ammar since its initial 
co~~ception by'Silvio Ceccato (Ceccata et al., 1960, 1963), 1 should like 
to dwell for a moment on a purely theoretical point. 1hile the term 
I I 
grammar" is predominantky used to indicate the formalized description 
11 
of a language (e.g. Chomsky, 1965; 4 and.140), correlational grammar" 
is, instead, the description of an interpretive system. The main difference 
between the two, though basically simple, has perhaps not been made 
sufficiently explicit. An ordinary gfammar is expected to account for 
all grammatical sentences of the language in a mare or less axiomatic 
way, i.e. by demonstrating that every possible grammatical sentence is 
a case under a formally stated rule or set of rules. An interpretive 
grammar, on the other hand, is not concerned with demonstrating the 
grammaticality of any sentence, but with transfoming the contept of a 
given piece of language into a canonical fbrh composed of pre-established 
semantic elements or modules, 
It is a "gr;immarl! in the sense that it 
consists of rules that govern this transformation, but these rules describe 
the language only indirectly, since what they actually describe is a model 
of the language user in the receiving role. (Mote that by "model", in 
-- --- 
this context, we intend a processor which, given the same input, will 
yield the same output as the processor to be modelled, regardless of the 
means it employs to do so. ) An interpretive system of this kind, thus, 
presupposes the grammaticality of its input. But since it is designed to 
interpret all grammatical pieces of language, it can be used to define 
f f 
operationally as grammatical" any input that it can interpret, while 
11 
input that it cannot interpret can be considered ungrammatical". 
When designing a correlational grammar for a natural language, it 
i5 a truly enornous problem to bring the gramarrs interpretive capability 
anywhere near the interpretive capability of the native user of the 
language. In the case of an artificial language, hailever, this problem 
is altogether eliminated, because the lexicon, the rules of concatenation, 
and the interpretive grammar can be designed all at the same titre. 
Since there is no native user, who has a universe of experiential content 
and well-established semantic connections (by means of which this 
content is linked to linguistic expressions), the designer is free to 
tailor the lexicon, as well as the syntax of his language, to the 
universe of discourse he envisages. 
That is, to a large extent, how Yerkish was designed, expecially 
with regard to the rules of grammar. The result of it is that the user 
of Yerkish can communicate in grammatically correct-lexigram strings no 
more than the correlational grammar of Yerkish can interpret. 
A Restricted Universe of Dis'course 
- - 
Yerkish, as it operates at present, is in fact a compromise In 
more than one respect. An effoft was made to create a potential universe 
of discourse that would allow a non-human primate to formulate as many 
communications as possible which, given the particular environment, 
could be used instrumentally for the attainment of goals (von Glasers- 
feld, 1974a). Such an attempt is necessarily based on more or less 
anthropocentric conjecture. There is, however, a certain amount of 
evidence that non-human primates organize their perceptual world in 
a way that does not seem incompatible with ours. In actual fact, 
Lana has already demonstrated that all the fterns which we assumed 
would take on the function of goals for her and would, therefore, act 
as incentive to communicatory activity, were indeed appropriate. Where 
food and drink were concerned, this could almost be taken for granted. 
h'ith visual displays such as a movie and slides, with thesounds of music 
and voices, and with the view through an open window, our anthropocentric 
hope of analogy was well rewarded. Above all it is gratifying t~ note 
that there was never a need to resort to any fom of negative r~inforce- 
ment or punishment. Though there were, especially at the beginning, not 
very mgny things that Lana could "say'' in Yerkish, she has never tired 
of saying them. 
On the practical side, since the interpretive grarmnar was to be 
implemented in a functioning parser, the universe of discourse was 
strictly limited by the size of the computer that could be obtained 
within the budget of the project. Becau'se the project is ~I~olly 
experimental and explorative, it was and is an absolut@ requirerent to 
leave within the computerized system a certain amount of room for 
ad hoc modifications and additions that might suddenly prove necessary 
-- 
in view of our subject's actual performance. 
Thus it was essential 
that the implemented grammar should never occupy all of the available 
space within the comp'uter. This is smll the case and we hope to be 
able to maintain this flexibility for some time to come. 
Technical Cons txaints 
There are four ways in which the Yerkish universe of discourse 
is restricted. First, there is the nunber of lexical Items the system 
can handle. The present version of the llultistore parser can deal 
with a maximum of 250 Lexigrams. The interface that links the con-.puter 
to the keyboard in the experimental chamber is designed for half that 
number, i.e. for 125. The-keyboard, however, is divided into five panels 
of 25 keys each and these panels are readily exchafigeable. This means 
that the subject's vocabulary can, in fact, be extended to 250 items, 
but only a subset of these, namely 125, will be operative during any 
one session. (Since Lana at present uses a total of 100 lexigrams, there 
is still much room for vocabulary expansion.), 
The second restriction also concerns the vocabulary of lexigrams, 
but it springs from the grammar of Yerkish and does not limit the number 
of individual lexigrarns but rather the number of conceptual classes to 
which lexigrams have to be assigned. Because of its interpretive 
function, correlational grammar requires a classification of lexical 
items that differs considerably from the word-classification used by 
txadi tional dcscr iptive gramars. Lexigrams , in fact, are classified 
according to certain functional characteristics of the concepts thev 
designate, i. e., according to cognitive characteristics. 
The lexicon with which a correlational gramnar operates, therefore, 
is divided, not into a few generic and largely rnorphological~y defined 
classes such as nouns, verbs, adjectives, etc. , but into a much larger 
number of classes defined in terms of what the designated items can do, 
i.e, , by the role or roles they play in the cognitive representation of 
experiential situations. In the case of "things1'- this is, for instance, 
the kinds of activity which they can perform as actors and the kinds of 
activity in which they can play the part of patient; and in the case 
of "activities" it is, for instance, the kinds of change they bring 
about. 
In the implementation of the interpretive system, i.e. the parser, 
it id the characterization of the lexical classes that occupies considerable 
space, not the individual lexical items. The total number of classes, 
therefore, has to be decided - a priori. In the present Yerkish parser, 
the maximum number of lexigram classes is 46. At the time of rkiting, 
35 of these classes have been filled (see Table 3). The remaining 11 
are still empty, but they can be made operative at any moment by the 
simple insertion of new lexigrams and the definition of the functional 
properties uf the items they designate, 
The third restriction concerns the number of lexigrams that can be 
strung together to form one message. The amount of data the parser has 
to take into account during the processing of a given message, obviously, 
depends to some extent on the number of lexical items of which the 
message is composed. This dimension corresponds to sentence length in 
natural languages. As it was impossible to foresee with any precision 
just how much work space the parser might require for the analysis of 
all types of grammatical input strings, we preferred to be on tile safe 
side and limited' sentence length to seven lexigrams. On the basis of 
the experience gathered since then, we can 13ow say that the computer 
system could, 'in fact, handle input strings of up to ten lexigrams and, 
hence, we plan to extend the capacity of the hterface hardware in the 
near future fr~m seven to ten Lexigrams. 
The fourth restriction involves the number of connectives (see 
Table 4) by means of which phrases and sentences can be put together. 
These connectives or correlatoxs are Ear more numerous in a correlational 
grammar than a?e the traditional syntactic functions. This proliferation 
is again the result of the interpretive purpose of the systen. A parser 
that is intended to extract the conceptual content from pieces of 
language must be able to idehtify not only the conceptual items involved, 
but also the relational- concepts by means of which they are connected 
with one another. Hence, the traditional distinction between syntax 
and semantics is no longer operative in a correlational grammar, and 
11 
the few basic grammatical relations" (e.g. subject-verb, verb-object, etc.) 
which connect grammatically characterized items, are replaced by a great 
f r 
many correlators" which are considered the linguistic expression of the 
relational concepts that link items on the conceptual level. 
While our English grammar operated with some five hundred correlators, (2) 
the grammar of ~erkish in its present implementation is limited to 46. Of 
these, 34 have so far been specified and are functioning (see Table 4). 
The remaining 12 will be filled as additions to the grammar became 
desirable from an experimental point of view. 
The Grammar of Yerkish 
- --. 
The interpretive purpose of correlational grammars leads to a 
shifting of focus from dharacteristics of words and sentences, qua 
linguistic items, to the- characteristics of concepts and conceptual 
structures, qua cognitive items. Ideally, a correlational grammar should 
be a complete mapping of the semantic connections between the elements 
and structures of a given language, on the one hand, and the elements 
and structures of conceptual representation, on the other. The bulk of 
vork required to produce such a mapping for a given natural language 
is so vast as to be almost forbidding. Nevertheless, work in that 
direction continues under various headings and significant advances 
have been made (e. g. Schank, 1972, 1973) . It will take a good deal 
more time to map the semantics of an average language user's universe of 
drscourse, but that is hardly a reason for not going on with it, 
expecially since much of what has been done encouragesthe hope that the 
task can, indeed, be completed. 
In designing an artificial language with a drastically curtailed 
universe of discourse, the problem is far more manageable. The semantic 
connections can be made as univocal as desired and, consequently, the 
process of interpretation can be thoroughly systematic. In the case of 
Yerkisb, unlvocality was desirable not only with a view to the- size of 
the automatic parser but also from the point of view of the teaching 
strategies to be employed with a non-human subject* Hence, Yerkish was 
(2) cf . Final Scientific Report , Automatic English Sentence Analysis, 
(December 1969) Grant AFOSR 1319-67, Georgia Institute for 
Research, Athens, Georgia. (Obtainable through D.O.D.) 
made as univocal as possible. 
Since both on the linguistic and on the c~nceptual level we are 
dealing with eiement~ and their concaternation in structures, the 
interpretive grammar has to specify the connections (a) between linguistic 
and conceptual elements and (b) between linguistic and conceptual struc- 
tures. With regard to the elements that are concatenated on the linguistic 
level, their semantic specification ca2 be given in the lexicon because, 
here, we are dealing with a fixed set of items, i.e, precisely, lexical 
items. With regard to structures - phrases and sentences on the 
linguistic level, and situational repr6sentations on the conceptual level - 
they have to be specified by rules of composition on concatenation, i.e. 
11 
by a grammar, because language is open" in that direction and allows of 
a practically infinite numter of individually different vord concatena- 
t ions. 
Because Yerkish is based on Lnglish and the output of subjects in 
the experimental environment will be evaluated by English speakers, 
the lexical semantics of Yerkish lexigrams c~uld be left implicit to a 
certain extent. Thus, for instance, the Yerkish parser does not have to 
contain an exhaustive semantic analysis of lexigrams such as BALL or 
RAISIN, because it can be taken for granted that the reader of the 
parser's output will be quite familiar with the concepts designated by 
11 
"ball" or by raisin" qua experiential items. What the parser must 
contain, however, is a mapping of- those specific characteristics of the 
concepts which determine these items' potential for entering into 
structural relations with other items. 
In Yerkish, then, the relational characteristics of conceptual items 
determine the classification of lexigrams . Thus, having decided, for 
instance, that there should be items that can be eaten and items that 
can be drunk, the lexigrams designating thss-e items will be divided into 
edibles (i.e. suitable patient/okjects for the activity designated by 
EAT) and drinkables (i.e. suitable patientlobjects for the activity 
designated by DRINK). Together they constitufethe class of ingestibles 
which, as it happens, is marked by the red hue of the corresponding 
lexigrams (see Tables 2 and 3). 
In short, Yerkfsh grammar does not require, nor lead to, a complete 
semantic analysis of lexical items. What it does require is a lexicon 
in which classes of lexical items are exhaustively characterized as to 
the specific relations into which their members can enter with members 
of other classes. This exhaustive characterization is supplied, not 
by listing all the other classes with whose members connections can be 
potentially formed, bat by a string of indices, each of which specifies 
a connective relation and the place in it (c-f. below) 5 member of the 
class thus characterized can occupy, 
Finally, we come to the relational concepts or coxrelators which 
are instrumental in the building up of complex structures, both on the 
conceptual and on the linguistic level. Strictly speaking, a correlator 
is a connective fgnction that links conceptual items on the cognitive- 
representational level. Languages indicate these connective functions 
by a variety of means: prepositions; verbs, nouns, and other types of 
words that incorporate a preposition; conjunctions and other particles; 
11 
syntactic markers" and, very frequently, merely word-order . Since 
these linguistic elements indicate correlators, we should call them 
I I 
correlator expressions". 
Rowever, once it has been made clear that 
correlators function on the conceptual level and connect concepts with 
other concepts or combinations thereof, we can in most cases use the 
term "correlator" for both the relational concepts and the linguistic 
devices that express them, 
(3) 
In designing an artificial language, the classification -of lexical 
items and the definition or explication of relational concepts must go 
hand in hand since the first is done in terms of the second, The rela- 
tional concepts have to be explicitly listed and explicated by some 
form of paraphrase. In principle, that is what a "case grammar" 
does. Its cases, basically, are relational concepts (e.g. Tillmore, 
1968). Ho~~ever, because correlational grammar attempts to cover as 
much relational semantics as possible, its list of correlators will be 
I I 
both much longer and wore specific than the lists of cases'' l~hich, 
to my knowledge, have been suggested. 
Yerkish; in its present form operates with some thirty correlators 
and the Yerkish lexicon is classified with reference to these (see Tables 
3 and 5). 
Given a basic list of correlators and their linguistic expression, 
the classification of lexical items can be carried out by listing 
for each item the correlators by means of which it can be potentially 
(3) One area where the distinction has to be maintained is the -semantic 
analysis of natural languages, because correlator expressions such as 
prepositions rarely have a one-to-one correspondence to relational 
concepts; instead, they merely mark the preqence of one of a set of 
relational concepts. 
linked to other items. To give an example, there is a relational, 
concept (No. U ) paraphrased as active ingestion of solids involving 
solid food stuff'; on the linguistic level, this correlator is expressed 
by the juXtaposition of two lexical items in a certain order. If we 
now have a lexigra EAT, that designates active ingestion of solids' 
and another lexigram RAISIN, that designates a subcategory of 'solid food 
stuff '., we can form a compound or correlation with the two lexigrms 
which can be represented as the structure: 
(a) EAT RAI S IN 
L,~~~-~,~~~~~-- 11 -- ------------A 
Because the order of succession of the two items in the linear 
linguistic expression is obligatory and cannot be reversed, it is not 
enough for the grammar merely to supply the information that the lexigrams 
EAT and RAISIN can be linked by correlator No. 11, but the grammar must 
aLss specify that, in this correlation, EAT has to be the left-hand 
piece (LH) and RAISIN the right-hand + piece (W). 
This information is part of the permanent lexicoq of the system. 
If 
It is recorded there by means of correlation indices" (IC1s), which 
consist of the number of the potential correlator plus the indication 
whether the items to which this I, is assigned can function as LH-piece 
or as w-piece. In many cases there are, of course, several lexical 
items that can function in the same place (e.g. NUT, MGM candy, RAISIN, 
etc., as RH-piece of correlator No. 11). Therefore , 1,'s are assigned 
to lexigram classes, not to single lexical items. Thus, while the 
lexigram EAT, in the present Yerkish lexicon, is the only member of the 
class VE ('active ingestion of solids'), the lexigrarr RAISIN is one of 
several in the class EU ('solid food stuff1). On the one hand, this 
indexing of classes, rather than individuals, is obviously more economical 
with regard t~ storage space, on the otKer, it makes it possible to 
add new lexigrams to the existing classes without in any way disturbing 
the operative part of the lexicon or the parsing algorithms. 
To expand the above example, let us add another correlation. The 
1 
relational concept that can be paraphrased as 
autonomous animate actorf 
perf ordng 'stationary activity' is correlator No. 01, The paraphrase 
! 
autonomous animate actor' comprises three lexigram classes in the 
present lexicon, nsmely kP ('familiar primates', i.e., the regular tech- 
nicians TIN, SHELLEY, BEVERLY, and the experimental animal LANA) ; AV 
('visiting primates', i.e., unnamed human or non-human visitors); and 
A0 ('non-primates', i.e., at present ROACH only). The paraphrase 
1 
stationary activity1, i.e, acti.vities that do not involve a change of 
place on the part of the actor, nor a change of hands on the part of a 
patient, comprises three lexigram classes, nanely VE (with the single 
member EAT), VD (with the single member DRINK), and LrA (with several 
members such as: GROOII, TICKLE ,. HOLD, etc. ) . 
Given the lexigran sequence LANA EAT, the interpretive grammar finds 
that LANA, belonging to class AF, bears the. I,: 
01, LH, while EAT, 
belonging to class VE, bears the I,: 01, RH; and on the strength of this 
the grammar will allow the correlation: 
LANA CAT 
"---------01---------4 
I I 
For the parser, allowing a correlation" means to record it as a 
possible part-interpretation of the input striqg.. As such it is recorded 
as a "product" in order to be tested for its potential correlability with 
other parts of the input. 
The information, on the basis of which such first-level correlations 
(connecting single lexigrams as in - a and b) are formed, is contained in 
7 
the permanent lexic~n and the form in which it is stored can be 
visualised as a kind of matrix (see Fig. 2 and Table 5). 
The correlational data required to form examples (a) and (bj is 
represented by markers (x) indicating the 1,'s (at head of column) 
assigned to the lexigxam classes (at beginning of row), In the present 
implementation of the Yerkish grammar h = 34, m = 35. 
Though this information contained in the lexicon covers all 
correlations involving two single lexigrams, it does not provide 'or 
cogrelations linking phrases or phrases and lexigrams. The syscem a 
correlational grammar uses to cfiscouer higher-level structures in a given 
input string is again rather different from that of traditional glammar. 
It order to be able to handle phrases, i.e., already correlated lexigrams, 
1 t 
or productsf', in exactly the same vay as single lexical items, each 
1 
product must be assigned a string of I c s that represents its particular 
potential for functioning as component (LH-piece or CH-piece) of a new 
and larger correlation that links it with other lexical items or phrases. 
The procedure that assigns these IC'S to a given product is what 
might be called the dynamic part of the grammar, because it is poverned 
by .an set of operational rul-es tha.t cannot be stated in- a siaple formalized 
way. (4) The reason for this is that the. correlruliility of a given phrase 
often depends on more than one constituent of the phrase. 
An example may help to make this clear. Wit11 regard to. rorrelator 
1'0, 30 that links the two single lexigrarris involved, the phrases 
and TIIIS CALI, 
L------30------1 
are identical. Ps potential YH-pieces of a correlation, forlllad by 
correlator KO, 11, however, they are not equivalent. 
EAT TEIS PAISIF 
1 
&------ ------.J 
30 
would be acceptable and correct, whereas 
EAT TNI S BA'I t 
I L ------A 
i 
30 
L----,, 
I1 
would not Ee acceptable because BALL does not belong to the lexipr,am class 
El' clef ined as !.solid food stuff.' and, therefore. is not a potential RH-piecc 
(4) The operational rules are, of course, always conFinations of indiv- 
dually simple rules taken from a relatively small set. This is, indeed 
the way in which the parsing program compi1.e~ them; although. this - can 
be called 'f,ormalisatj-on" it. certainly is not a simple one. 
of correlaticln No.. 11. In fact, if the string EAT THIS BALL occurs as 
input to the interpretive grammar, it must he rejected as incorrect. 
To implement this dl~criminat ion, the phrase TIIIS MISIP! must be 
assigned the I,: 11, EH, while the phrase THIS DAI.IA must not. And in 
order to do this, the assignation must be based not only on the 
particular correlator that links THIS with another item, but also on the 
condition that t'his other item is one that belongs tu the lexigram class 
solid food stuff'. In other vords, there has to be an operational 
assignation rule that makes sure that a first-level correlation produced 
by correlator KO. 30 is assigned tbe Ic:ll, PiT, so that it can I-e linked 
in a second-level correlation with the precedirp lexigram FAT, v~hich 
bears the I 11, 1.1;; but this a'ssignation must. be made contingent upon the 
c 
condition that the product 30 (P.: 30) does, in fac.t, contaic a lexical' itern 
of class El' as RH-piece; because only if P:30 contains a member of tl~r: 
t 
class solid food stuff' can it function as patient of the activity 
designated by the LH lexigram FAT. 
The operational assignation rules, therefore, are of diverse Lypes, 
I ? 
sore assigning I, s unconditionally, others assigning I, s only on condi- 
tion that the same Ic is present, as the case ray Fe, arrong those charae- 
terising the 1A1' or the 911 of the produet thgt is tieipg classified. (.See 
Table 6). 
1 
In the implementation of the parser., tile assignation of I, s to 
products is primarily dete'mined by the particular correlator that is 
involved in the product to be classified, Tk.e assignation rales a particu- 
lar correlator calls into action, tliouph functionally of three types 
only, are specific. to that correlator and cannot be written in a generalised 
form; 
This indeed, is the fundamental reason why a correlational grammar 
cannot be represented by means of a small number- of relat-ively "powerful" 
rules. 
In a correlational grammar there must be as many sets of 
specific 
assignation rules as there are correlators ; and since the. number of 
correlators in such an interpretive gram.ar is very much laqger 
than the 
number of "syntactic functions" in conventional descriptive grammars, 
correlational grammars connot be wrj tten in concise and powerful formulas. 
As a justification for this lack of Ebrmal elegance, however, it can be 
said that correlational grammar has no need of the otherwise indispensable 
and somewhat unwieldy adjunct of "selection rules", because it incorporates 
that very information in its one basic interpretive algorithm, 
Peculiarilies -- of the Yerkisk Gr.mar 
The grammar of Yerkish had to be kept as simple as possible for the 
reasons mentioned above, First, given the small size of the computer, 
it was mandatory to avoid complex constructians and rules of grammar that 
might require special space- and time-consuming subroutines in the 
parsing pfocedure. Second, the rules of the language to which the lingpi- 
stic behavior of our subject would have ta confo;rm, were to be few and 
consistent from the learner's point of view; nevertheless they were co be 
such that Yerkish structures could be translated easily and without major 
structural transformations into comprehensible English. As a result of 
these objectives, Yerkish grammar may seem somewhat unusual. 
In the fol- 
lowing paragraphs several deviations from English grammar will be dis- 
cvssed. 
Yerkish, at present, bas only one voice. the active, and three moods, 
i. e. indicative, interrogative, ,and imperative. Both :he interrogative 
and the imperative are formed) not ,by specific v,erb-forms or word-order 
(as in ~lilny natural languages), but by sentential prefixes or markers, i.e. 
specific lexigrams that are placed at the beginning of the message. The 
prefix of the interrogative is the conventional question mark 'I?"', that 
11 
for imperatives (in Yerkish requests") is an arrow, translated into 
English as PLEASE. The keys representing these two lexigrams nust be 
pressed at the beginning of a string and they can appear only in the first 
feedback projector on the left. The lexigram string following them has 
the same form as an .indicative statement. In fact, if the string is to he 
interpreted as an indicative-statement, i.e.. if it is - not preceded by 
!I 711 
either . or PLEASE, the first feedl;acl;projector on the left remains 
blank. Thus: 
TIM MOVE INTO RpOM = indicative statement; 
? TIIIE YOVE INTO ROOF? = interrogative; 
PLEASE TIE1 MQVE INTO ROOM = request. 
A third lexigram that fu.nc-tions as a sentential marker is NO, which 
corresponds to an over-all negation of the statement. NO TIM MOVE INTO 
ROOM, therefore, corresponds L-o the English "it is not the case that 
Tim moves into the room". However, since Lana has quite spontaneously 
come to use the lexigram NO to designate what, given the situational 
context, could be interpreted onLy as "don't", we may adapt the grammar 
to her usage and turn this NO into a marker for negative imperatives (see 
Appendix) . 
Yerkish, as yet, has no tenses but the present. A simple past and 
future, however, are foreseen, and they will be designated by particles 
preceding the activity lexigram. 
There are no auxiliaries in Yerkish and the function of the English 
copula "to be" is taken over by corr,elator No. 10, which is expressed by 
juxtaposition of a lexigram belonging to one of the classes of items that 
1 t 
are modifiable" and a lexigram designating a specific state. 
e.g. BALL RED. 
- 
"the ball is red1'; 
T IEI AWAY , 
- 
1 I' 
Tim is away"; 
The absence of an explicit copula is noticeable also in conjunction 
with the "naming functionf1, an important instrument in ~ana's acquisition 
of new lexical items. It is used for the ostensive definition of new 
lexigrams which areplaced at the beginning of a string of the form: 
XX NME-OF THISi 
- 
11 
XX is the name of this1'. 
(where LY is the new Icx~gram) 
Two English coQstructions that have a specificatory restrictive 
function., i. e. for instance, "the red ball'! and "the ball is red" 
are one and the same in Yerkish, and the specificatgry relation is 
designated by a lexigram which can be translated into English as the 
comp~und 'WHICH-IS (correlator )lo. 31) . 
e. BALL ITHICH-IS RED. "the red ball" or "the bal 
which -is red1'. 
For the sake of greater univocality, Yer,kish spatial prepositions 
were strictly divided into locatignal and directional ones. The first -- 
e.g. IN, ON,OUTSIDE, etc. -- could designate ~nly the locatign of items or 
activities, the second -- e.g. INTO, OUT-OF, FROM, etc. -- could designate 
only the direction of activities involving a change of place. Ho~lever, 
since Lana has spontaneously used a locational prepositYon to indicate 
the target of a directional activity, and since this is allowable in 
many if not all natural lartguages, we are considering the removal of this 
restriction with regard to spatial prepositions. 
So far, there are no conjunctions in Yerkish, but a somewhat restric- 
ted form of "and" and "or" has been worked out and will shortly be intro- 
duced into the system, 
There are also some minor pecularities that an English-speaker must 
keep in mind.. A Yerkish structure involving correlacor No. IT, for 
instance, implies that the speaker is the receiver of the item that 
changes hands, unless another teceiver is explicitly indicated by a pre- 
positional complement. Thus, if Lana sends the message : 
PLEASE TIME GIVE IIILK. 
it must be understood that the milk is to be given to Lana. The receiver, 
however, can be made explicit by adding a prepositional phrase, which 
yields the correlational structure: 
PLEASE TIM GIVE MILK TO LANAP 
-17--1 224 
-21- 
Lh0L5 I 
It 
English resultative verbs, e.g. to open", "to clean", ete., are, 
broken up in Yerkish. The causative element is rendered by EIAKE, the 
effect by a lexigram designating the resulting state. Also, in Yerkish 
the agenr must be specified. Thus,  l lease (Timj open the window1' 
becomes : 
PLEASE TIM MAKE WINDOW OPEN. 
Translated literally into English, this should be  lease, Tim, make 
window - be open", since the correlator that links WlNDOW and OPEN is No. 10, 
i.e. the gredicative copula equivalent to "to betf. But in this case, as. 
indeed in most occurrences of correlator No. 10, the Yerkish string is 
easily understood without the explicit copula. 
The Yerkish MA~E is not limit-ed to causation of a change of state of 
specific items, but can be used also to indicate a number of perceptual 
conditions or events in the environment. Specific sensory events or 
changes, such a NOVIE, NUSIC, SLIDE, HEAT, COLD, LIGHT, and DARKNESS, are 
considered the- result* of activities subsumed by .RAKE. In Lana1 s wholly 
technoPogica1 environment. this is not at all unreasonable. It obviously 
makes sense for her to request: 
PLEASE PLACYINE MAKE MOVIE. 
L13 --L 
L.16 2 
L40 - 
It is, indeed, the machine that causes the projector to start running. 
S2milarly, however, in Yerkish one cauld correctly say: 
PLEASE TIM MAKE LESS HEAT. 
It 
Though in Lana's expe~ience Tim can indeed cause less heat' by turning 
down the thermostat, 
this would hardly he a reasonable request in the 
"realIt world outside the Yerkes Lab. 
MAKE also opens the way to embedded constructions. since it can 
govern a clause. Though ,Lana has not yet come to this, the grammar foresees 
strings such as : 
? TIM NAKE LANA SEE VISITOR 
or even a double embedding: 
? TIM SEE LANA MAKE ROACH MOVE. 
and similar structures are, of course, possible with WANT. 
Lest these correlational diagrams create the impression that Yerkish 
structures are invariably right-branching, here are two examples that 
contain lef t-branchings : 
? NO PIECE OF APPLE HERE. 
which,in English, would read': "IS there no piece of apple here?" And 
STICK WHICH-1s BLUE DIRTY 
-10 - 
which in English, would be: "?he stick which is blue is dirty", and, as 
such, roughly equivalent to "The blue stick is dirty. 
11 
Irr one particular the grammar of Yerkish deviates from correlational 
practice. 
Prepositions and conjunct ions being "explicitH 'correlator 
expressions in that they designate relational c.oncepts only, are (in the' 
correlational approach) n~t items-to be linked, but itenla that do the 
linking. Thus, in the original Ffultis tore parser they functioned as 
cor~elators and not as ordinary lexical items. In the structure diagrams, 
therefore, they appeared in a node, not at a terminal. Given the very 
smLl computer used for the Yerkish parser; as well as the fact that the 
lexicon was t'o remain extremely limited (in comparison to English), it was 
more economical to correlate prepositional phrases in two steps rather than 
introduce the special routine that had been developed for prepositions and 
other "explicit" correlator expressions in the English pagser. Thus a 
string. such as "move into room" is not constructed as it would be in a 
proper correlational systan, i.e.: 
but rather in twa steps: 
NOVE INTO ROOM 
1 
L-4 
where P:21 containing P:22 expresses the conceptual relation designated by 
INTO. 
In all other respects the Yerkish system is similar to the Multistofe 
parser whose characteristic data-compression was, in fact, the feature 
that made .it possible to contain t.he entire system - lexicon, operational 
interpretive grammar, and automatic response programs - in less than 5000 
machine words of central core. 
Conclusion 
Though Yerkish is, indeed, an extremely limited linguistic sys tem, 
the examples of sentence structure I have used above should -be suffi'cient 
to show that it has a considerable range and flexibility with regard to 
what can be formulated in it. The reports on ~ana's progress that we 
have published so far (Rumbaugh et -* a1 9 1973b, 1974; von Glasersfeld, 
1974b-; see also Appendix to this paper) leave little doubt that Lana has 
already acquired a number of skills that certainly belong to what is 
usually called linguistic competence. The grammar of Yerkish as it is 
at present allows many structures which are still far out of Lana's 
reach. She has a long way to go before one might venture to say that 
she fully exploits the expressive possibilities of Yerkish. That is 
precisely how it was intended at the outset of the project. 
In any case, the range of expression could easiLy be extended at 
short notice and without interfering with the existing operational system. 
Nor would such additions require an inordinate expansion of the 
Lexicon. The introduction of the one lexigrarrl WANT, for instance, has 
opened the way to a completely new level of expression that may even- 
tually lead to a demonstration of the chimpanzee's capability for concep- 
tual representation. The addition of a Yerkish "if. . . then'' would be 
no nore difficult and could pe,rhaps further clarify the cognitive poten- 
tial of non-human prinates. (5) Such additions could also make Yerkish a 
(5) Premack (1971) reported that his chimpanzee Sara could correctly 
interpret an "if. . . then" connection between actual activities 
or states; the greater range and flexibility of Yerkish would make 
possible the introduction of much more sophksticated hypothetical 
statements. 
valuable co~unication vehicle for sone of those many unfortunate 
children wh6, though they are mentally not at all deficient, remain 
averbal because of some physiological damage. It is towards this end 
that we are now exploring the possibility of adapting the Yerkish syst,em, 
its grammar, and the parser, to a form of simple English. 

References 
Brown, Rog.er (1971$, in Ploog and Melnechuk, 1971. 

Ceccato, Silvio, -- et al. (1970) , Liriguistic Analysis and Progr.amming for 
Mechanical Translation, Nlan, Italy: Feltrinelli. 

Ceccato, Silvib, et al. (1963), Mechanical TranslatLon: The Correlational 
Solution (Technical ~e~ort), Milan, f taly : Center for Cybef netics , 
~nivk'rsity of Milan, 

Chomsky, Noam (1965), Aspects of the Theory of Syntax, Cambridge, Massa- 
chusetts: M.I.T. Press. 

Crawford, James (1973), personal comunication. 

Fillmore, C. J. (1968), The case for case, in (E. Bach and R. Harms, Eds .) 
Universals in Linguistic Theory, New York: Holt, Rinehart, & Winston. 

Gardner, R. A. and B. T. (196?), Teaching sign language to a chimpanzee, 
Science, 165, 664-672. 

Gardner, B! T-. and R. A. (19.71), Two-way communication with an infant 
chimpanzee, in (A.M. Scihrier and F. Stollnitz, ~ds.) ~ehavior of 
Ndnhuman Primates, Vol. 4, New York: Academic Press. 

Hayes, Keith J. and Nissen, C. H. (1956),, in (A.M. Schrier and F. Stollni-tz, 
Eds.) Behavior of -Nonhuman Primates, New York: Academic Preps, 1971. 

Hockett, Charles F. (1961), Lin-uistiq elevents and their relations, 
Language 37, 1, 29-53. 

Ktlhler, ~olf~ael925), The Mentality of Apes, New York: Harcourt , Brace. 

Lenneberg , Eric c1971) , in Ploog and Melnechuk, 1971. 

Lieberman, P. (1968), Primate vocalizatbon. and human linguistic ability, 
Journal of the Acoustic. Society of America,'4, 1574-1584. 

Lorenz, Konrad (1974), Analogy, as a source of knowledge, Science 185, 
229-234. 

Lycns, John (1'972.), Human language, in (R.A. Hinde, Ed. ) Non-Verbal 
Comunication, Cambridge, England : 
Cambridge University Press. 

Ploog, Detlev, and Melnechuk, Theodore (1971), Are apes capable of language?,. 
Neurosciences Research Bulletin, 9,, 5. 

Premack, David (1971), On the assessment of language competence in, the 
chimpanzee, in (A.M. Schrier and F. ~tollnitz, Eas.) Behavior of 
Nonhuman Primates, Vol. 4, New York: Academfc Press. 

Bumbaugh, Duane PI. (WO), Learning skills of anthropoids, (in L. A. 
Rosenblum, Ed.) Frimate Behavior, Vol. 1, New York: Academic Press. 

Bumbaugh, G; I., von Glasersfeld, E., Warner, h., Pisani, P., Gill, T., 
Brown, J., and Bell, ,C. (1973a), A,computer-controlled la~guage 
training'system 'for investigating the language skills of young apes, 
Behavioral Research Ilethods and Instrumentation 5, 5, 385-392. 

Bumbaugh, C. t!., Gill, T. V., and von Glasarsfe-1d;-T., (l973b) , Reading 
and Sentence completion by a chimpanzee, Science 182, 731-733. 

Rumbaugh, D. M., von Glasersfeld, E., Warner, R., Pisani, P., and Gill,.T. 
(1974), Lana (thTmpanzee) learning language: A progress report, 
Brain, and Language, 1, 205-212. 

Schank, Eager (1972), Conceptual dependency: A theory of natural language 
understanding, Cognitive Psychology, 3, 4, 552-631. 

Schank, Roger, (1973) , Causality and Reasoning, Technical report ii~. 1, 
Castagnola, Switzerland: Fondazione nalle 1.iolle. 

vdn Glasersfeld, Emst '(1964) , A project for automatic sentence analysis, 
Beitrsge. zur sprachkun4P und Inforrnationsverarbeitung , 4, 38-46. 

von Glasersfeld, Ernst (1964); Efultistore - a procedure for correlational 
analysis, Automazione,e Automatismi, 9, 2. 

von Glasersfeld, Ernst (1370), The correlational Lapp~oach to language, 
Pensiero e Linguaggio, 1, 4, 391-398. 

von Glasersfeld, '.~rnst (19 74a) , Signs, communication, and. language, 
Journa'l of IIurnan Evolution, 3, 465-474. 

von Glasersfeld, Ernst (1974b), Lana's progress, paper presented at the 
12th lleeting of the Association for Computational Linguistics, Amherst 
l!assachusetts, July, 1974. 

von Glasersfeld, Ernst, and Pisani, Pier Paolo (l970), The'?iultistore 
parser for hierarchical syntactic structures, Cormnu~ications of the 
Association for ~orn~utin~' Elachinery , 13, 2, 74-82. 

von Glasersf eld, ~rnst, and Pisani, Pier Faolo (1968), The hfultis tore 
Sys tern PIP-2, Scientific Progress report, Athens, Georgia: 
Georgia Institute for Research. 

Yerkes, Robert 11. (1925), Traits of young chirpanzees, Cn (R. .?.I. ~erkes 
afid B. \J. Learned, Eds,) Chimpanzee Intelligence. and its Vocal 
Expressipn, ~altkore : Williams & ~il"1iams. ' 
