1. education
2. language learning

Chapter 1
What Is Language?
When we study humon language. we ore approaching whot
co" rhe "humon essence," the distinctiue qualifies 01
mind that ore, so for as we know.'E"ni,quefo mafi1
NQ/lm Chom,ky. L.c"9VOSlt o!1Cf Mlld
some might
doll011Y Hart
B.C.
1 l.~'='~~c.o
A NE,'t' u,.~
'if) "" Ole WIlL "'''''''''
,., ru.<J~& APO-,.r. «,ur
By pcrminion
W>JJr 1'0"
f>\Tf,"l'" ~.
(G\J I\'EJ<C
,"~S'.~D
o-E.
t/1"~frlE 0..0
or Johnny Hart .f'Id Fit:ld Enlcfl'ns.c", In( .
.'
.;
Whatever
else people may do when they come to.
gether-whether
they play. fight. make love. or make automobiles-they
talk. We live in a world of words. We talk to our friends. our associates, our
wives and husbands. our lovers. our teachers, our parents and in. laws. We
talk to busdrivers and total strangers. We talk face to face and over the telephone. And everyone responds with more talk. Television and radio further
swe\lthis torrent of words. As a result, hardly a moment of our waking lives
is free from words. and even in our dreams we talk and are talked to. We
also talk when there is nO one to answer. Some of us talk aloud in our sleep.
We talk to our pets and sometimes to ourselves. And weare'tlle MIy-:rnimals
~hat oOlnis:':thatlalk1
,..
.- - .J
.The possession of language. more than any other attribute. distinguishes
humans from other animals. To understand our humanity one must understand the language that makes us human. According to the philosophy eXpressed in the myths and religions of many peoples,iillslangiiage--Ihal"isthe
{so~rce"orhumanlifeand-poV;'ei'1 To some people of Africa. a newborn child
is akunru. a "thing," not yet amunlU. a "person." Only by the act oflearning language does the child become a human being.' Thus according to this
tradition. we a\l become "human" because we a\l knoW at least one language. But what does it mean to "know" a language?
, Diab3te. Mas ••.
• . "Ora! Tradition and Mali Literatu«:'
Mak n
(Mali Information Cenler).
in Tht Rtp.bric
of Moli
3
,!
:'\1
4
The Nature of Human Language
Linguistic Knowledge
\:
When YOUknoW a language. ¥oUC1UIS1)taK'and lJeiin~
(st\foll-15y-o!ners,wno knowthat-Ianguage. This means you have the capacity
to produce sounds that signify certain meanings and to understand or interpret the sounds produced by others. We are referring here to normal-hearing
individuals. !)eaf persons produce and understand sign languages just as
hearing persons produce and understand spoken languages.
Everyone knoWS a language. Why write an entire book on what appears to
ld
be sa simple a phenomenon' After ,all. five-year-o children are alm~l as
proficient at speaking and understanding as are their parents. YeUhe abilitY
to-carry- out"tlll!'Simplesl conversati()l\-requi res 'profound _knowledge .that
'Sl5ll\kerSare 'n-a r<:of. This is as true of speakers of Japanese as of Eriu ofwa
glis speakers.
EskimoS as of NavajoS. The fact that we may knoW something
unconsciouslY
is not unique to language. A speaker of English can
h
produce a sentence with tWOrelative clauses like
My goddaughterwhOlives in Sweden is named Oi••• which was lhe name of a Yilt-
!
I
,
I
I
I
!
i
I
"
i
I,
,
I
\
\
I
,
I
I
;
I
i,
I
lng queen.
I
without knowing what a relative clause is. This is parallel to Imowing hoW to
wall<.without understanding or being able to explain the neurophysiological
I
control
that
permit
oneknoW
to doEnglish
so.
What. mechanisms
then. do YOU
knoW
i(YoU
or Quechua or French or
I
e
Probably
Moha
Arabic' being aware of it. you knoW ih SOu'nClsthat are part of
wlt or without
your language as well as those that are not. This knowledge is often revealed
by the way speakers of one language pronounce words from another language. 11 yOU speak only English. for example. you may (and usually do)
substitute an English sound for a non-English sound when pronouncing
"foreign" words. HoW many of you pronounce the name Bach with a final k
sound? This is not the German pronunciation. The sound represented by the
letters eh in German is not an English sound. \f yOUpronounce it as the Germans do. yOUare using a sound outside of the English sound system. Have
yOUnoticed that French people speaking English often pronounce words like
Ihis and rhat as if they were spelled ,is and tar? This is because the English
sound represented by the initial lellers Ih is not part of the French sound
system. and the French mispronunciation reveals the speakers' unconsciouS
Kno ••••.
ledge
of the
knowledge
of this
fact.sound patterns of a language also includes knowing
which sounds may ~artawo1<l. end--award• and ~<>lIow'each'other. The
name of a former president of Ghana was Nkrumah.-Ghanaians pronounce
tl)is name with an initial sound identical to the sound ending the English
word sing (for most Americans). But most speakers of English would mis.
pronounce it (by Ghanaian standards) by inserting a short vowel before or
after the n sound. Similarly. Ngaio Marsh. the Australian mystery story
••••.
riter's first name. is usually mispronounced in this way. There is a good
reason for these "errors." No word in English begins with the ng sound.
Children ••••.
ho learn English discover this fact about our language. just as
Ghanaian and Australian aboriginal children learn that words in their language may begin with the ng sound.
_ ,
•••••••••••
"
••
1
I
Whal Is Language'
\
\
5
Knowing the sounds andSQuliifpaifeliiS1in one's language consliluteS only
one part of our linguistic knowledge, A most important part of knowing a
language is knowing that certain sounds or sound sequences signify or represent different concepts or "meanings," That is. if you know English, yoU
know lhat boy means something different from lOY or girl or plNOdoCI)'1.
Knowing a language is thereforel1<'iiOWinlrthe'System"lhanelates"SoundS'1lnd
,m'e'llning$1 If you don'l know a language. the sounds spoken to you will ~
pretty much incomprehensible. This is because the relationship belween
speech sounds and the meanings they represent is. for the most part. aii3r6i~trafy-one, You have to learn (when you are acquiring the language) 'thattt{e
'sounds represented by the letters IlOllS( (in the written form of the language)
r
\
~
t
signify the concept
-62t: if you
know French. this same "meaning"
is
represented by maisan; if you know Twi, it is represented by ,daQ: if you
know Russian. by dam; if you knoW Spanish, by coso.
Similarly, the concept
is represented by hand in English, main in
=::3
French, nsa in Twi. and ruka in Russian.
The following are words with definite meanings in some different languages. How many of them can you understand?
I
i
\
\
a. KyinH
b. doakam
c. odun
d. asubuhi
e. toowq
f. bolna
g. wartawan
h. inaminalu
i. yawwa
If you don't know the languages from which these words are taken. you un.
doubtedly don't knoW that they mean the following:
a. a large parasol (in a Ghanaian language, Twi)
b. livingcreature (in an American Indian language. Papago)
c. wood (in Turkish)
d. morning (in Swahili)
e. is seeing (in a California indian language. Luisenol
f. to speak (in a Pakistani language. Urdu): ache (in Russian)
g. reporter (in Indonesian)
h. teacher (in a Venezuelan Indian language, Warao)
i. right on! (in a Nigerianlanguage. Hausa)
These different words show that the sounds of words are only given meaning by the language in which they occur. The idea that something is called X
because it looks like X or called Y because il sounds like Y waS satirized by
Mark Twain in his book Eve's Diary:
The minute I set eyes on an animal I know what it is. I don't have to renect a mOment: the right namecomes oul instantly....
1seem 10 knowjusl by the shape of
the creature and Ihe way it acts whal animal il is. When the dodo came along he
[Adam] thoughl it was a wildcat. . , . But I saved him....
I just spoke up in a
quile natural way ... and ",id "Well. I do declare if there iso'tlhe dodo~"
No mailer what one's opinion of Eve's wisdom as compared to Adam's. it
is clear that neither the shape nor other physical attributes determine the
I
I
:
6
The Nature 01 Human Language
sounds or n3mes of most cre3lUres or objects in any langu3ge, as is so
cle3r1y shown by the Herman cartoon. A pterodactyl could have been called
Ron.
"I. Vagel'
HERMAN
G
WE NEED SOME
MOI1f 1IXlRn5 FOR
;a OUll IJ>,NGUAGE.
'7
WAAT II~E WE GOIIlG 10 CALL
tI'J'oT 6ta 81Rl>.1HING WI1H lIlE
lEl(1\4E~ WING~ -;>
V-
.
c::r
AWft
PTERODACT'I~ ?
PlERODAtm!
RON!
~
\llE1L AT lEAST"
PEOPLE Will BE II111E
TO SPELL IT !."
@ 1981 Universal Pte\!'. SyndICate, Reprinted .••..
ith pcnni!i.~on. All riShis (C~,....ed.
i
I
\
\
This arbitrary rel3tionship between the form (sounds) and meaning of a
word in spoken language is also true of the sign languages used by the deaf.
This is easy to prove. If you watch a sign interpreter on television with the
audio turned off, it is highly doubtful that you will understand the message
being conveyed (unless of course you know American Sign LanguageASL-or
Signed English). A nonhearing user of Chinese Sign Language
would also find it difficult to understand a user of ASL. Many signs. of
course, may originate as visu31 imit3tions of their referents: they m3Y be
mimetic (similar to miming) or iconic (with a nonarbilrary relationship be.
tween form and meaning) to begin with. But signs change historically as do
wordi, and the iconicity is lost. These signs become conventional; in the
same sense that knowing the sounds of words does not reveal their me3ning,
so knowing the shape or movement of the hands does not reveal the meaning
of the geslUreS in sign languages.
-
_._-------------_ .................•••••••
•
What Is Language?
II
;ThUS: ti,e
conv.ntionar and-al'bitrary'.nature.'of
7
form:..meaning-relation-
Isliipsin languages-spoken
and sign~is universal1
- There is, however, some "sound symbolism" in language. That is, there
are words whose pronunciation Suggests the meaning. A small group of
words in the vocabulary of most languages is "onomatopoeic"-the
sounds
of the words "imitate" the sounds of "nature." Even here, the sounds differ
from one language to another, reflecting the particular sound system of lhe
language. In English we say cockadoodledoo and in Russian they say ku-
I
kuriku
the rooster's
One to
alsorepresent
finds particular
sound crow.
sequences that seem to relate to a particular concept. In English many of the words beginning with gl have to do with
sight, such as glare, glint, gleam. gliller, glossy, glaze. glance, glimmer.
glimpse, and glisre . Many rhyming word pairs begin with h: hoiry-toil}'.
n
harum-scarum. hotsy.totsy, higgledy-piggledy. But these are a very small
part of any language, and gl may have nothing to do with "sight" words in
When language.
you know English you know these gl words, the onomatopoeic
another
words, and all the words in the basic vocabulary of the language. you knoW
their sounds and yOU know their meanings. It's extremely unlikely, of
course, that there are any speakers of English who knoW the 450,000 words
listed in Webster's Third New International Dictionary. But even if they did,
and that was all they knew, they would not know English. Imagine trying to
leam a foreign language by buying a dictionary and memorizing words. No
matter how many words you leamed, you would not be able to form the simplest phrases or sentences in the language or understand what was said by a
native speaker. No one speaks in isolated words. (Of course you could
search in your traveler's dictionary for individual words to find out how to
say something like "car_gas-where~"
After many tries, a native might
understand this question and then point in the direction of a gas station, If
she answered YOUin a sentence, however, it is highly probable that yOU
would be unable to understand her or even look up what she said in your
dictionary, since you would not know where one word ended and another
\
\
\
\
\
~
Your knowledge of a language enables yoU to 'combine-words "to-form
began.)
Olhrases;-and-Phrases-tO"fOrm Sentence~ UnfortumitelY, you can't buy a dictionarY with all the sentences in any language, since no dictionary can list all
the possible sentences.%,nowing ilanguage i\\eans-being able-to produce
neW"'"ftntllfces never-spoken before and to understand sentences never
heard before: The linguist Noam Choin'sky refers to this ability as part of the
~'creative aspect" of language use. This doesn't mean that every speaker of
a language can create great literature, but it does mean that you. and all persons who knoW a language, can and often do "create" new sentences every
time yOU speak and are able to understand new sentences "created" by
others. This is because language use is not limited to stimulus-response behavior. We are "free" from the constraints of either internal or extemal
events or states. If someone steps on our toes we will "automaticallY" respond with a scream or gasp or grunt. These sounds are really not part of
language; they are involuntary reactions to stimuli. After we automaticallY
CTY out, however, we can say "That was some clumsY act, you big oaf" or
"Thank yoU very much for stepping on my toe because 1 was afraid 1 had
elephantiasis and now that 1can feel it hurt I know it isn't so," or anyone of
L
\aI.
'iht
F,
,-
~e
\01
lbe
\;.
~do
jh.
ag,
\
J••••••••••••••••••.•................
I'••••
---.--------------
8 The Nature of Human Language
ar
an infinile number of senlences. because lhe oarticul senlence we produce
is nol
conI rolled
any in,oluntary
slimulus.
AClually.
e,en bysome
cries are conslrained by our own lan-
0""
••,m.",
,. "ri'.'"
".~,' """''';'','
"',,' "_.;.,
" " .'"" ",,'" '"",,"",w "";, E'.;"
_ ,.,,' ",,,i' """ l~,d
\
\
t
only in Ihe language. French speakers. for example. fill Iheir npauses
wilh Ihe
,owel sound Ihal slarts Iheir word for egg_Oell! _and does ' occur in Enes
Of words
course or
knowing
glish
cries ora language
paus - also meansl"nowlng wliatsenrencesarea'ii-
•••••••
".;,'~ri''''
""."••••
;",i,," H,m""'" ,,, •• ".00 ,,,,,,, ..
afler someone has just stepPed on your toe during a discussion on lhe
wealher in Brilain would hardly be an appropriale response. but il would be
\
i
\
\
I
I
\
.,I
possible.
Consider. for example. the following sentence:
Daniel Boone decided to become a pioneer because he dreamed of pigeon-loed
. giraffes and cross-eyed elephants dancing in pink skirts and green berelS on Ihe
l
wind-swepl plains of Ihe Midwes .
YOIl might not belie,e the senlence: yOUmighl queslion ils logic: yOUmight
e,en understand ittO mean differentlhings: bul yOUcan undersland Ihe sentence. allhOllgh it is ,ery doublful Ihal yOUha,e heard or read il before noW.
II is ob,ious. then. that when you knOWa language you can recognize and
undersland and produce neW senlences. All of them do nol ha,e 10 be as
"wild" as Ihe Daniel Boone senlence. In facl if you gO through Ihis book
counling the number of senlences you ha,e e,er seen or heard before. we
predict Ihe number would be ,ery smal\. Nexllime yOUwrite an essay or an
exam or a leller see hoW many of your senlences are new. \l can'l be Ihal all
\
\
",,,,,,, ,,",""
,m ,,~, ;, ,~, bmi,,,' .10'w'"
,~
""k ,~ ","
out a sentence which seems 10 fit Ihe silualion. or Ihat when you hear a ,enlence you malch il with some senlence already stored. HoW can one ha,e in
his or her memory a tOlallY no,el senlence ne,er heard before7
In facl. it can be shown Ihal simple memorizalion of all the possible sentences in a langllag is impossible in principle. If for e,ery senlence in the
e a longer senlence. then there is no limit on Ihe length
language one can form
of any senlence and therefore no limit on the number of senlences. We can
il\lls
Ihis by a well-known example in English. When you knoW Ihe lantrate
gllage, you knoW yOUcan say:
\
\
\
\
This is Ihe house.
or
This is the hOuseIhal Jack buil\.
or
This is Ihe mall Ihal lay in Ihe house Ihat0'lack buil\.
This is Ihe dog Ihal chosed lhe cal Ihal killed lhe ral Ihat ale lhe mall Ihal lay in
Ihe house Ihat Jack buil\.
And one needn' stop there. HoW long.lhen• is Ihe longest sentence7 One
t
,
j
t,
can also say:
-,
,
.\
••••••••••••••••••••••••••••••••••••
What Is Language?
I
\
The old man came.
iuce
9
or
The old, old, old, old, old man came.
How many "old's" are too many? Seven? Twenty.three?
We will not deny that the longer these sentences hecome, the less likely
one would be to hear or to say them. A sentence with 276 occurrences of
"old" would be highly unlikely in either speech or wriling, even to describe
Melhuselah. But such a senlence is theoretically possible. That is, if you
know English, you have the knowledge to add any number of adjeclives as
modifiers to a noun, as is illustrated in the Wizard o[ Id cartoon.
Dr.at Parker aDel JObDDY H.rt
THE WIZARD OF 10
~
••••••• .;:a....-o l,.'''. ~
~-";..6 pt(~
~
"00':7
lI'~l,..•.••.
;.vo:;. ~<r~.::-...'o'"
..",••~T'
F";;:.,~'0001
Alfl!l'.:::-Ji"
~
.,py' •••~
:,::)otl".
~
: •••• FI.:~T
~I(A"v~I-,'i>:'"G
~r'
CO'''
"7.•(.~l,.,,",".~""e.l!"4,..e-~-:'"'~
e.t" •••.•. ":'",•.•..-,
A...:;!
e'oJO.?"I","'G,AOlCI
A1.~ ••••
G""<".1'
By pcrml~~on of Johnny Hart and Field Enterpris.es.
!toI'''':''''
...•• "TfP
oF tJe~
•.
Inc.
To memorize and store an infinite set of sentences would require an infinite slOrage capacity. But the brain is finite, and even if it were not we could
I
not slOre totally novel sentences.
But when you learn a language you must learn something, and that something must be finite, The vocabulary is finite (however large it may be), and
that can be stored. If sentences in a language were formed by putting one
word after another in any order, then one's knowledge of a language could
be described simply by a list of words. That this is not the case can be seen
by examining the following strings of words:
\
t,
I
(1) a. John kissed the little old lady who owned the shaggy dog.
b. Who owned the shaggy dog John kissed the liltle old lady.
c. John is difficultto love.
d. It is difflCultto love John.
e. John is anxious to go.
f. It is anxious to go John.
g. John who was a student flunked his exams.
h. Exams his flunked student a was who John.
If you were asked to put a star or asterisk before the examples that
seemed "[unny" or "no good" to you, which ones would you "star"?' Our
"intuitive" knowledge about what ";s" or "is not" a good sentence in
English convinces us to "star" b,f, and h. Which ones did you "star"?
I
11 has become CU'iIOmaryin presenting lingui':ilic data 10 use the aSlerisk before any ct.-
ampl•• lhal ,peake<>rejeelfor one reasonor anolher.We,hall use Ihi, nolalionIh,oughoUI
thebook.
,
••i,••••..............
------------.
10
The Nature of Human Language
Would yOUagree with our judgments about the following?
(2) a. What he did was climb a tree.
b. -What he thought was want a sports car.
c. Drink your beer and go home!
d .• What are yOU drinking and go home? t
e. I •• peet them to arrive a week from n•• Thursday.
f. 'I ••peet a week from next Thursday to arrive them.
LinuS lost his security blanket.
h .• Lost linuS security blanket his.
g.
If YOU':starred" the same ones we did, then it is clear that not all strings
of words eonstitute sentences in a language, and our knowledge of the
es addition"to
language determines which do and which do not. Therefore. ~n
<knowingth.-"words -onn€langu.age yb(jmu~CKnow- some"i'ul '
to fonrt,
the-senlences and to make the judgments that you made about the examples
in (I) and (2). These rules"mustoe "finite'in length and"filiite"in number SOthey
can be stored in our finite brains. Yet they must permit uS to form and understand an infinite set of new sentences as was discussed above. HoW this is
words;
possible
discussed
Chapter consists
7.
We canwill
saybethen
that a in
language
of all the'-s(jifn~.
and
tlossible "Senterices. And when you knoW a language yOUknoW the sounds.
the words. and therules-fonneifcomblnaiion~
"
What You KnoW and What You
Do: Linguistic Knowledge and
Performance
"What's one and one and one and one and one and one and
one and one and one and one?"
'" don't know," said Afice. ""ost count"
"She can't do Addition," the Red Queen interTlJpted.
t;zv.'6 earroU. Through tht Loobng.GIas.S
Charle, Schlitz
PEANUTS
@ 1%4 United
Fc2lu1C
Syndicate.
InC.
We have mentioned some aspects of speakers' linguistic
knowledge such as the ability to form longer and longer sentences by joining
sentences and phrases together or adding modifiers to a nOun. We also
pointed out that such sentences are theoreticallY possible. but hardly practica\. Whether one limits the number of adjectives to three. five, or eighteen in
speaking. it is impossible to limit the number one could add if desired. This
demonstrates that there is a difference between having lhe necessary knowledge to produce sentences of a language and the way we use this knowledge
._------------_ ..........•.•.. ,
What Is Language?
11
in linguistic performance or behavior. It is a difference between what one
knows. which some linguists refer to as one's linguistic competence or capacity. and how one uses this knowledge in actual behavior. which we can refer
to You.
as linguistic
performance.
as a speaker.
have the knowledge to understand or produce very
long sentences (in fact. as noted above. no limit can be set on the length of a
sentence in any language). But when you attempt to use that knowledgewhen yOUperform linguistically-there
are physiological and psycholOgical
reasons why you limit the number of adjectives. adverbs. clauses. and so on.
You may run out of breath; your audience may leave; yOUmay lose track of
what has been said if the sentence is toO long and overloads your short.term
memory; and. of course. yoU don't live forever.
When we speak we have a certain message to put forth. At some stage in
the act of producing speech we have our thoughts organized into strings of
words. But errors occur. We all produce speech errors or "slips of the
tongue" such as the one in the Wizard of ld cartoon.
WIZARD OF ID
S
By pcnnis~on of Johnny Hat' and FiekS Entcn'ri5C
,
Inc.
Such errors also show the difference between our linguistic knowledge and
the way we use that knowledge in performance.
In discussing what you knoW about your language it is important to repeat
that much of your knowledge is not conscious. You learn the linguistic system-the
sounds. structures. meanings. words, and rules for putting them
all together-without
anyone teaching them to you and without being aware
that you are learning any rules at all. lust as we may be unconscious of the
rules that alloW us to stand or walk. to crawl on all fours if we choose. to
jump or catch a baseball. or to ride a bicycle. our unconscious ability to
speak and understand. and to make judgments about sentences reveals our
knowledge of the rules of our language. This knowledge represents a complex cognitive system. The nature of this system is what this book is all
about.
What
Is Grammar?
We use the lerm "grommor" with a systematic ambiguity. On the
one hand. the lerm refers to Ihe eJ<PficitIheOI\lconstructed by
the linguist and proposed as a description of the speaker's
competence. On the other hand. [II rejers Ito this competence
itself.
N Chomsl<Y and lot
II
\
.-----_.-
Holle. The Sound P•••••• 01 £ngbh
7
.I
The Nature ot Human Language
12
R
d
GRAMrotA
I,""
,
I,"""
'0' I"m,h, ""d, ,w i' ,••' I"W"" ,I<
••'i' "i" of ~"",. ",h " •• ,d'. "d 'l< m ,••
\h'"
D£SCRIVTI'/£
Wl<'><"
'0 ~."'
••
\0
fOl. ,oW "'"....
'"
01'."" ,od ",,, ,,,,, i"" "" ", •• " ,f •
I", ••'" '" ,~"d""'"
•• '" ,1<"i,
.h" wo\"""
I""i"i',••
,', ,,""
,f ,,,,,,,, iI"'''"''''''
" .'"h "d",,,,d
""" ,f ,", i",,m" i"d. 'OI00"iOO'", of"I•• ,hil i, ,," ,f ",'
E'''' "."
""i"
," ,,"'<, ,,,,,,,,,oo~ \h' •••• ". Wl<'I•.
grammar
of everY
language.
.
••"" ",h '0 d'"'''''' , '"',''' ,h" "".,' \0 d'''''''' ,h' ", •• " of
,', I"'"'' ,hoI"i'" " ,l<.i,d' ofi" ,,,.''''' "'" ." ,f ~,~ "
,,~ """"",
h" "d \h" ,f
""h". 8'" h'" "'",,,,
.'" •• 'l<
,h",d ,,,,"d,, ••,"" di, \h"
", •• " ,hoI
.,,'" i' p""ib ,fOl,",'''' \0 ,,1\ ,,,",, "d",,,,d ,01 .", l<'.• , \h'
l
",,", d," ,l<Ii",''''''
d'"'''''' " , w' """'01 of,l<"""" Ii",''''
'" ",,,i'" d ,ill "" ••• ,' " ••dd,"",IiOOof,l<,,, •• " ,f ,I<I'"
,,, •. "d of,l<I" ••••.•d"If. S••h, """''' i, "I\W d~",Ii" "'~
•••• \I d",0" ,,11 ,,, hOW,,, ,h" Id 'I'"''
d "',,''''''
,," ,,,"
1i,••"'iI ,,,,,,d." d",,,i" h" di, ""ib" '"' ,,, \0 ,,,.' "d ".
",,,,,,d. "d iI ","''' ,h" d " ,,, "OW .,," 'l< ""d, .•• ,d'.
phrases,
sentences
of your
W,M"and,,,d
,l<.,,"',
••language.
" "IW' •• " ,l<,,,, i' "f'"~'"
"•
•••••
,,"''''
,
••
'
i,
,l<" ,~i."
'l< ,~,"" " ,I<""",,, 0< d~"~
,iI",f "" '"" ""Ii'''' Y' •• '" A1~" ,., ,,,,,,,,,",,, ••••• ' ,MG"'"
~.,' "," .0
'00'''''' ".,'" "",,,
.",h
"" •• '"'' Oioo',,,, "". d""d ,~ •• " " ,•••'
,,,.i" ," "
""" ,p<" , I~""" 0< ,p<'" .1""• I"" •••.f ,"" ~, 0' " ,ill,"
d~,""' "" ,M" ,~ ~~i"'. ,i.' ,l<Ii""i"" d'"""'" ,~ •• '" i'
~ "".,'
" • f,"""
,u".'" (0< ,,..,N) 01 ,l< ",,''''
••• ""'" ".
".,h""
••, "h"b'",••
\h., \h'" i,• ~I,i' \h' ,~ •••• "" ."
..,,'N ~,,,""
• ,~, ,""
,h"'"
"';0" "d • ,,<I> ,h'"" ,,,,,,,,,."
,", i, ""i"d " • ~" i, ,,,h ,M..• ,,,." ,,~.'" "d <h' _I ,f ",I<Ii",i""
•••••. A"",'" " ••, \h" • ","," '''" •• ,,'''' "
~" \h" iI i, f_ " ,,,,'"' , ••' ,l< "", of ;o<h ,,, •• '" ~,.
d
,,~I,
.." '"'"••••,,'
''''''"''~,,,'"d",,'"i' ,,~ ." I'''''
,M~
~I". If."OW""." ,,,iI • ~" fo< E""iI' ,hi,h ,",' ,,, •••"
WI'" ,i" ~,
\h'
.'~.'"
"
."
d'''""'"'
"
i, -'
••, "ff,"" I'''''
,l<,,,,,.'"
.,,,h","~'" ",,,'hi'"
,,,, I"""'"If.",,,,,.
~.P<",,,..,M'".,," ,,,.
,~" i,
,h'" " • .i",,"•••
•,"
M"
0" ",~",i"
,
,,
•
•""
A""""
,M~••
, of ,," Y'."'"
"'" "ff" f". ", ~'" of_~"' ,,~., ,,, ••••.. ,l<" ,,,., P"""" M
,ii' ,," i" ","," " • ""."" ,M' ,l<" i, "."""
,~."'.
0<
00' ,',
,"
••
,"
•• i"'" i, ,," •••••
.~o<""'"Ii","'"
"i, "'~~"
I~'
,,'" ~ ••,i," ,f • I"'"'' ,,,,II'" • d"",,1 " "",,0< " ,,' ",,, " •
Ii"'"'" ~'" E"N ,,,•• '" i, "',." '0."" "d IOS".1 ", ""."'
,I ,,,,,,,,i" " ""iI' ,,'of ",",'"
'0 ","" '" "",,,,,, ,", .,.'"
I ~
,','
" ","'"
If "."""
"" l< """,,'
I~'. iI"" l<","",d " '" "l<' I"'~"
,,,~,."" "d d~"~'
~,d,.• " il'"
\
" ,,,
,,,,~".
0' ".,,,,.
'\
;
.y\
~
•
0<0" d••
Y" .,,'" ,~ d,f.
8~"""~.
,M"",,,,'" ~ •••••
"" "
"""'""""
,f
preferred except perhapS for nonlingui,tic reasons.
•• , ., , •••' d",,,,,i~ <h'~""
\..
00
What Is Language?
,
13
t
\
PRESCRIPTIVE GRAMMARS
"I don't wanlto lalk grammar, I wanlto tolk uke 0 lady,"
G 5, S~W. Pygt'"ohon
,the
, to
or a
islic
itbC
,crY
,\in~of
ebe
to!
\!Iat
,tbC
pis'
\aDlID"
!Sic
un-
~,
'\he
rip:ek
,to
)Ot
lis
\a1
IS:
,"
,
ie
•m
tt
\If
1Il
The views that are expressed in the section above about descriptive
grammars are nottho,e of all grammarians noW or in the pas\. From ancient
times until the present there have been "purists" who have believed that
language change is corruption and that there are certain correct forms
which all "educated" people should use in speaking and writing, The Greek
Alexandrians in the first century, the Arabic scholars at Basra in the eighth
century. and numerous English grammarians of the eighteenth and nineteenth centuries held this view, They wished to prescribe the roles of grammar rather than described the roles. Prescriptive grammars were therefore
Il-
it
u
It
lll
It
'1\
~
{.
With the rise of capitalism and the emergence of a neW middle clasS. there
written,
•t
\'
\
What Is Language?
15
lects are expressive, complete, and logical and as much so as they were 200
or 2000 years ago. If sentences are muddled this is because language is a
powerful tool for expressing one's thoughts and because some speakers'
performance abilities may be lacking. Prescriptivists should be more concerned about the thoughts of the speakers than the language they use.
C-
l,e
/\h,
oDe
red
J1b
trY
~:.
"Hopefully"
thisofbook
will convince
you of this.
When we talk
the grammar
of a language
we are also differentiating the
notion of grammar from teaching grammars, which are used to help speakers
learn another language, or even a second dialect. In some countries where
it is economically or socially advantageouS to speak a "prestige" dialect,
people whO do not speak it natively may wish to learn it. Teaching grammars state explicitly the rules of the language, list the words and lheir pronuncialions, and thuS are aids in leaming a neW language or dialect.
In this book we shall not be primarily interested in either prescriptive or
teaching grammars. We shall, however, discUSS the question further in
Chapter 8 when we discuSS standard and nonstandard dialects.
\(Or
Language Universals
In a grammar there are ports whichpertain 10 all languages;
the•• componentsform whaCis coiledthe general grammar....
In addition to these general (universol)ports, Ihere are those
whichbelong onlyto one porticularlanguage; and these
constitutethe porticular grammars of each Ionguage.
Ou MarWs. c. 1750
The way we are using the word grammar differs in an-
olher way from its most common meaning. In our sense, the grammar includes everything speakers knoW about their language-the
sound system,
called phonology, the system of meanings, called semantiCS, the rules of word
formation, called morphology, and the rules of sentence formation, called
syntax. It also of course includes the vocabulary of words-the
dictionary or
lexicon. Many people think of the grammar of a language as referring solely
to the syntactic rules. This lalter sense is what students usually mean when
they talk about their class in "English grammar."
Our aim is more in keeping with that stated in 1784 by the grammarian
John Fell in "Essay Towards an English Grammar": "It is certainly the
business of a grammarian to find out, and not to make. the lawS of a language." This is just what the linguist attempts to do-to find out the laws of
a language, and the lawS that pertain to al/ languages. Those laws that pertain to all human languages. representing the universal properties of language, constitute what may be called a universal grammar.
Throughout the ages, philosophers and linguists have been divided on the
question of whether there are universal properties that hold for all human
languages and are unique to them. Most modern linguists are on the side of
the "universalists,"
since common, universal properties are found in the
grammars of all languages. Such properties may be said to constitute a "universal"
human language.
About grammar
I630, theofGerman
philosopher Alsted first used the term general
grammar as distinct from special grammar. He believed that the function of
ageneral gramma~ was to reveal those features "which relate to the method
\,
\
16 The Nature of Human Language
and etiOlogy of grammatical concepts. TheY are common
a to all languages."
pointing out that "general grammar is the pattern 'norm ' of every particular
grammar whatsOever," he implored "eminentlinguists
to employ their inThree
andmatter.'"
a half centuries before Alsted, the scholar Robert Kilwardby
sight
in this
held that linguists should be concerned with discovering the nature of Ian.
guag in genera\. So concerned was KilwardbY with universal grammar that
he excluded considerations of the characteristicS of particular languages,
e
which he believed to be as "irrelevant to a science of grammar", the mao
terial of the measuring rod or the physical characteristicS of objects were to
geometry.'" In a sense, KilwardbY was 100 much of a universalist, for the
,.rt""~' "",rt;" of;od;
,;, ••1I•••••
~'., "I"'" W ,',
,;""",
•
language uni,ers , and are, in addition, of interest for their own sake.
The emphasisalsthese scholars placed on the universal properties of Ian.
guag may lead someone attempting to study Latin, Greek, French, or Swa.
hili as
e a second language to assert, in frustration, that those ancient scholars
were sO hidden in their ivory towers that they confused reality with idle
speculation. Yet the more we investigate this question the longer the list of
"universals" grows. The following list is far from complete but it gives us an
idea of some universal facts about human language. Some are factS about
language in general, and others refer to specific characteristicS and proper.
\
iI
,,
I
\
ties of the languages of the world.
\.
exist, language
exists. language are equally com2. Wherever
There are humans
no "primitive"
languages-all
plex and equally capable of expressing any idea in the universe. The
vocabulary of any language can be expanded to include new words for
\
I
\
neW conceptS.
3.
change
thrOUgh
4. All
Thelanguages
relationships
betWeen
thetime.
sounds and meanings of spoken Ian.
guag and between gestures (signs) and meanings of sign languages
es
for the languages
most part utili~e
arbitrary.
5. are
All human
a finite set of discrete sounds (or gestures)
that are combined to form meaningful elements or words, which them.
\
\
,
selves form an infinite set of possible sentences.
6. All grammars contain rules for the formation of words and sentences
7. of
Every
spoken
language includes discrete sound segments, like p, n,
a similar
kind.
or a, which can all be defined by a finite set of sound properties or
features. Every spoken language has a class of vowels and a clasS of
"
"
8. consonants'
Similar grammatical categories (for example, noun, verb) are found in
9. all
There
are semantic universals, such as "male" or "female," "ani.
languages.
mate" or "human," found in every language in the world.
10. Every language has a way of referring to past time, the abilitY to ne,
gate, the ability to form questions, issue commands, and SOon.
, V. Salmon. review of Cu,ltsian
Lin8uiS/irs by N. ChomSky, 10u,nu' of Lingui,'ir
(1%915: \65-IS7,
SiSn
langU.gop. of
•• V.
Salmon.
citlhe
. deaf do nOl, of course, use sounds. They are discussed in alaI
es
s~tion.
What Is Language?
11
II. Speakers of all languages are capable of producing and comprehending
an infinite set of sentences. Syntactic universals reveal that every language has a way of forming sentences similar to the following:
Linguistics
is an interesting o;ubjecl.
I \<no that linguistics is an interesting subjec\.
You \<no
w that I \<nowthat linguistics is an interesting subjec\.
Guineverew \<no that yoU \<nowthat I \<now that linguistics is an interesting
ws
w
Is
it
a
fact
that
Guinevere
\<nows that you \<now that I \<no that linguistics is
subject.
Qt;Jy
\lll'
that
p.
an interesting subject?
1lI1'
12. Any normal child. born anywhere in the world. of any racial. geographical. social. or economic heritage. is capable of leaming any language to which he or she is exposed. The differences we find among
,\0
\bC
1of
languages cannot be due to biological reasons.
\aD'
\l seems thai Alsted and Du Marsais (and we could add many other "universalistS" from all ages) were not spinning idle thoughts. We all speak
~
\alS
IIle
"human
to!
\311
'anguage."
In the Beginning: Language Origiil
llUt
lCf'
God created the
world
by a Word. instantaneous/Yowithout wil
and poins.
1hC
Talmud
Nothing.
nO doubt. would be more interesting than 10 know
from hislO
documents lhe exed process by which the firs!
man beganrical
to lisp his firs! words. and thus to be rid for eLlerof
all
lhe theories on the origin of speech.
to\ Mull«. 187t
The universalitY of language as a unique characteristiC of
e
the human animal also led to the question of hoW 'anguag originated. All
religions and mytholOgies contain stories of language origin. Philosophers
through the ages have argued the question. Scholarly workst have been
written on the subjec\. Prizes have been awarded for the "bes answer" to
this eternally perplexing problem. Theories of divine origin. evolutionary development. and language as a human invention have all been Suggested,
Such widespread speculation is not surprising. Man's' curiositY about
_himself\ed to his curiosity about language. Many of the early theories on the
origin of language resulted from man's interest in his own origins and his
own nature. Because man and language are SO closely related. it was believed that if one kneW hoW. when. and where language arose. perhaps one
".or
Ii
t-
a
would knoW hoW. when. and where man arose,
• tn Englishand in manylOla"') other languag". the masculineform, ofnoun' andpronouns
are used as the general.or generic. term. WewouldhaveliKedto avoidIhisbut foundourselves
constrainedby commonusage. Had we ,aid "Woman'SCuriosityabOuthersel{\ed10 hereuriosity abOutlanguage." thi, would have been intert'reted as referringonly to women.U,ing
larlhat the
word "man" in lhis scnle
.and other senle"" throughoutthe booK.we are sure
the
inlert'relationwillbe "mannCC
and woman." Wherever "man" or "mankind" or a ,iml generic
lS
term is used. the readeri, asKedto con,ider Ihese genera!lermsembraCInglhewholeofhuman'
ity. unless of course the meaningcan ,pecificallybe relaledto lhe male membe of the species.
,
18 The Nature 01 Human Language
The difficullies inherenl in answering these questions aboul llanguage are
immense. AnlhropologistS think Ihal man has existed for al leas one million
years. and perhaps for as long as five or six million years. But the earliest
and
deciphered wrillen records are barely six Ihous
years otd. daling from
Ihe wriling of Ihe Sumerians of 4000 B.C. These records appear 50 lale in the
history ofs the developmenl of language that Ihey provide no clue at all to the
i
\
One ofmighl
conclude that the quesl for Ihis knowledge is doomed 10 failure.
origin
language.
The only hard evidence we have aboul ancienl languages is writlen. bul
speech precedes wriling hislorically by an enormous period of lime. and
even loday Ihere are Ihousands of speech communilies speaking perfeclly
"uP'lo.date"
languages Ihat lack wriling systems. The language or languages used by our earliest anceslors are irretrievablY lost.
For Ihese reasons. scholars in Ihe latler part of the nine lee nih cenlury.
whO were only interes
in "hard science,"
ridiculed. ignored. and even
led
banned discussions of language origin. In 1886. Ihe Linguislic society of
Paris passed a resolulion "outlawing"
any papers concerned with this subThis ban was reconfirmed in 1911 and was furlher supported by Ihe presijecL'
denl of Ihe PhilolOgical SocielY of London. Alexander Ellis. whO concluded
in his address 10 the SocielY thai:
...
We shall do more by tracing the hislorical growth of one single work-a' day
longue. than by filling wastepaper baskets with reams o( paper covered wilh speculations on the origin of aU tongues.
Thai such resOIUlions did nol pUI an end 10 Ihe inlerest
facl Ihat jus a few years ago Ihe IinguiSI John P. Hughes
l
is clear from the
fell compelled to
wrile:
...
a word or IWOshould be said in any serious linguislic work to counter Ihe
arranl nonsense on this subjecl which is still circulated in Sunday supplement science features. According to this pseudo_evolutiOnary foolishness. based on nothing
but rampant imagination. language originated among our caveman ancestors when
someone tried to tell the hithertO speechless tribe about the wol( he had killed. and
was (orc to give an imitation o( the wol( ...
or when he hit his thumb with the
mallei while
sharpening
a
stone
spear.
sO
that
ouch became the word for
ed
"pain" ...
and similar fairy stories.'
This view shat1'ly diverges from Ihat pUI forth tWO hundred
by Lord MonbOddo.
Ihe Scollish
years earlier
anlhropolOgisl:
The origin o( an art so admirable and so use(ul as language ... must be allowed
to be a subject. not only of great curiosity. but likewise very important and interesl'
ing. i( we coosider. that it is necessarily connected with an inquiry into the original
nature o( man. and that primitive slate in which he was. before language was invented ...
.'0
, La S",i/'/ MUJ •• " ""'u., ,,, ••,nuni(l'I'''. ,,,.,,,non'."
rari~in' du '0.Ra~' .. :. r'
Society dotS not accept any paper concernin'
the origin of lanS86
on
The
guage ... ") La Societe de Llnguistique. Secll 1. Statutsllm ).
uoX
, lohn P. Hughes. t969 Th, s,i,net "f LanR"oX' IRando House. ~ew
Yor".)
in
" lames Burnell. Lord ",onbUddo. Of ,h, Orix unJ Prax"" of Lanx ' (\174).
--
.. - - .- ..
,
:'.\
"
;'
~,
• ,<,,,,,N ,,,di<'~I" '" ",." ,,'~,,""d.'." ",d.1,,,,.""
wOO ." >orod"" ,,,ro '" "" ,,,,.,,,,
•• ",
"."."
",,,. ,,'
" •• '" " .' ",,,,,,'
,,,, •• ,'" A,d d,,1 ,;;"'"
",,'" "," ,,,,.,," ,,'" "",'"
'" "", "w",~'"" ,••,'"
,;;!d'" ,. - w''''.
1',,"""
".j""".,,"N ,'" " ,',,,,,,.,"'" .1",,""
relates
to evolutionary
in the
brain. ,.. ,,,,.'" ,,~, "d i< "
"'OO''''
•• •• "lUchanges
'" Iro"
,~w,,'
being taught-bY
,1<" ""mere expOsure.
d""I.' ,'"
••
w'" ,," " •• I" "rt.", ••
'''' " •• ,,'" I" '"
,I< """
.1""" ,.".""
\
w'" " ,••" •• ,,,,,,,,,
\
",,'" ",d' ".' ""',," " ••,,,",,""
w, '"
,Il ,,,'ro"'" I.roll'"
V" I,. of"'
"" ,,"," ",,;dO' ."" •• ""
"",' "" """"
•• ,d' .1•• ,,"'"
.1<' •• '~"
"'" ••••.•1<"" ~
E,."'"
,.",,,"
"',.,,,•.••
o.'" """,
" • dl<''''''''' ,"' ~,
"",,>",..,,,,,'"
"d •
,I<
.1I ",
•••
,"" ,,,",,,,1",. "",,,,, " ••Il" ••"" .•""'" ••••" ""
.Il"",,,'" •••• ' "" ,I< wk' , ",,'''' ."' ro,rod.' ,oJ """,,,d
w'"
., ••
ld".'" "d '" ••" I••,. ,I< ""d ,,'"'' .1 '"w'",,,,,,,,
"d ,I< •• ,' " .,f<' ".oJ' "d "'" "" '" "ld"d"I<
_.,,'" ,. '" """d, ••• "'"''''
."d> '" ,,'''''' ,•• , ","""
I"''''''
.,"
",
'!y" "d ,,,,, ".,,1< ••'""', ••'" w"" ~" "
,,, "" ,,,,d', ,~ •• ," i< ro"'" "",",," ,1<'," • ,,'"'' "" ".
,_,0"
",e".,sentences.
••• 1., •••••• ,. V•• k'ro ,I<"
anTh<"
infinite set of
"pOssibl
'" k' ,I<
"I< ,,,",,,,,,,
ro
of
'"'' ,,,'"
.i<' ",,,'"'' ." .," '" "
;y •• 1I".",f< _",
System.
,,, "", 1I".",f<
E'" if ,.,
,.,d " "" ull'" '" ", • roPl"" " •• "
"" ,,,wl,d,,
1""""""'"
•• • "",.'" '" ,,'"
"
lro" ,." "'''''•• " ., .~
~d d'.
d••• " ", Y'00' • " •• - .1
"If'"''
".ro'"
,'k"''')' yo' .,," "ii','" ,,,,,,,,,,, 01, •• ,~,,,,,. "', ,bili""
"", ••,,, •• ",,1<' Ii"""" ",b"." " ,,, "d" ,,,' '" ,~,.""
"'" ".""
"".', Ii"' if ,,, " •• ,I< ,~ •• "" "'" "., " •••• "" "
'" U~-'01• ,~,.,,' ",,,,,,,, ,M 1i".Il'" ,~."d"" aspeak.
",ii, .1,,, ,,,"""' )l"".d" ,I< Ii"" ••"d', ."d" "d "k>'" ,M
l.~" •• ,"" ••""•• ,~d."~,,a,'o,
,I"." •••" Li"""" ",.1'
"',' ",,,."
"',,"'"'
S,,' • ,""_
,,,"'''
,d •• ,I< .,,,,,,,,
" ." ".,,''''
,~-"""
.• , """,
of '"" _, Ii"""" ,,,wkd<' " ","" • do"'""'" U''''''''
.1 '"
" • ",,,'
I"","'.
01,I< " ••• a'" ,ro"""
,~ •• " ""
k
,,~b ,', w ' .f ,I< ,,,,.,,.
iI d'
)l •••• ",
,,,,.Id b'
","" • "'",'.'''
,,, •••••,)l",,,'b<" " d"'" dOl''''', " ••,' ••
d"ol".•,~"" ,,,,,,,~ ." .,••,0'''' " ,01, ""k I"" , I",
language or a dialecl of their language that differs from their own.
TI<"," Ii"'"'' ,,,,"'",,',',\Ii",~d' of I~,",," .1"" •• M .
A,~"" "" """,,, " ,,,,,I'"~.,.,
scribes Ihe rules Ihat are already known.
yo. ,~""
dOl'0Ii' " ••• Y'••• ,," ,MY""" I"" ••"
"Ii", \Ii, ",,, ""
What Is Language?
29
\
I
f.
\,
,
I
'll'ans
i.and
cover that these differences are limited. There are universals of language that
pertain to all parts of linguistic grammars. The difference between universal
or general grammar and specific or special grammars has been the concern
of grammarians and philosophers throughout history.
our curiosity about ourselves and our most unique possession, language,
has also led to numerous theories about language origin. There is no way at
present to "prove" or "disprove" these hypotheses. but they are of interest
jdren
i\bOut
for the light they shed on the nature of human language.
The idea that language was God's gift to mankind is found in religions
throughout the world. The continuing belief in the miraculous powers of language is tied to this notion. The assumption of the divine origin of language
stimulated interest in discovering the first primeval language. There are legendary "experiments" in which children were isolated in the belief thattheir
first words would reveal the original language. Children will learn the language spoken to them: if they hear nO language they will speak none. Actual
cases of socially isolated children shoW that language develops only when
~e
ol us
there
is sufticient
linguistic
Opposing
theories
Suggestinput.
that language is a human invention. The Greeks
believed that an ancient "legislator" gave the true names to all things.
Others have Suggested that language developed from "cries of nature," or
"early gestures," or onomatopoeic words, or even from songs to express
iI nO
!be
111&
Jist
~
There is at present a renewed interest among biologists and linguists in the
love.
question of language origin. VariouS evolutionary theories that are now proposed oppose both the divine.origin and the invention theory. Rather, it is
suggested that in the course of evolution both the human species and language developed. Some scholars Suggest that this OCCUlTedsimultaneouslY,
and that from the start the human animal was innatelY equipped to learn language. In fact there are those whO believe that it is language which makes
human nature human. Studies of the evolutionary development of the brain
provide some evidence for physiolOgical, anatomiC, and "mental" precondi-
!leU
llF
(!be
lIl1
,by
ltc-
'ibis
I
~v-
tions for language development.
!e-
o!
~
ExERCISES
\. Part of your knowledge of English includes knowing what sound sequences occur in the language. When neW products are put on the market
the manufacturers have to think up neW names for them. and these names
must conform to the allowable sound patterns. Suppose YOUwere hired
by a manufacturer of soap products and your job was to name five new
products. What names might yOUcome up with? List them.
We are not interested in the spelling of the words but in hoW they are
pronounced. Therefore, describe in any way yOUcan hoW the words yoU
list should be pronounced. Suppose, for example, yoU named one soap
powder Blick. You can describe the sounds in any of the following ways:
a. bl as in "blood,"
i as in "pit,"
ck as in "stick"
b. bli as in "bliss," ck as in "tick"
c. b as in "bOY," lick as in "lick" and sO on.
j
1