Stops and other sound-symbolic devices expressing the relative length of referent sounds in onomatopoeia.

1. Introduction

Phonetic symbolism or a symbolic connection between the sound shape of a word and its referent has seriously attracted attention of many linguists since Sapir's "Study in phonetic symbolism" was published in 1929. Sapir's study started off a long series of experiments whose aim was to assign specific symbolic meanings referring, for example, to shape, color, brightness and size to sounds, based on native speakers' intuitions. The subjects of these experiments were either confronted with artificially created words or with words coming from various, often exotic, languages. Their judgements were then treated as a basis for assigning specific symbolic functions to individual sounds. The sound [i], for example, was found to appear in English words connoting small size and [a] in the ones that referred to big objects (Sapir 1929). Jespersen (1933) in his contribution to the search for the universality of sound symbolism found many confirming instances of the vowel [i] occurring in words carrying the meaning of "sm all, insignificant, weak and rapid" (Brown 1958: 118) in several Indo-European languages. It wouldn't be fair, however, to claim that there are no opponents to assigning sound-symbolic values to individual speech sounds. They even quote counterexamples to the observed regularities (cf. Brown 1958: 119). It is important to realize, however, that the rules of phonetic symbolism do not apply regardless of the environment in which a given sound-symbolically charged unit (feature, sound, morpheme) is set. As this paper will attempt to show, the phenomenon of phonetic symbolism is a multilevel one with a hierarchy of sound-symbolic devices, which depending on their patterns of cooccurrence in the same environment seriously affect each other.

Onomatopoeia being a class of words designed to imitate all sorts of sounds that may be heard in the world surrounding us is by definition a very abundant source of data for linguists looking for sound-symbolic connections between words and their referents. The phonosymbolic elements in onomatopoeia, however, will be less likely to connote such properties as size, color and shape but will rather be restricted to imitating sounds or the manner in which they are produced. 0-word corpora will prove to be a real goldmine for those researchers who expect to find symbolically charged sounds/sound classes among those that show patterns of overrepresentation (when confronted with lexicon samples) when subjected to a sound frequency analysis (cf. Sobkowiak 1990; Kwiatkowski 1992; zuchowski 1993). The main focus of this paper is on a single aspect of phonetic symbolism in onomatopoeia, namely the somewhat previously neglected sound-symbolic function of stops. The study will be performed on three corpora of o-words; Eng lish, Polish and French. A hypothesis will be tested in which it will be claimed that o-words containing stops tend to refer to short and abrupt sounds and actions, whereas those onomatopoeic words which do not contain stops tend to refer to prolonged or gradually occurring sounds and actions. It will also be claimed that the stops' ability to perform their sound-symbolic function may be seriously affected by other phonosymbolic devices.

2. The data

The empirical data used for this paper consists of three corpora of onomatopoeic words: the English corpus of 250 o-words originally compiled by Sobkowiak (1990) based on Kloe's English-Spanish dictionary of onomatopoeia and other sources, the 191 -item Polish corpus compiled by Kwiatkowski (1992) and the author of this paper, based on Polish language dictionaries, comics and poems, the French corpus of 146 items, most of which were acquired by Sobkowiak from The University of California, Berkeley and later on made available to me for my research.

3. The phonosymbolic role of stops in English, Polish and French onomatopoeia

It was suggested in several works on sound symbolism that stops appearing in onomatopoeic words carry the notion of brevity and abruptness of the action referred to (cf. Leech 1969; Brown 1958). To my knowledge, however, no one has made an attempt at testing this hypothesis based on the whole corpora of o-words. Instead, linguists restrict themselves to providing single examples, which although indicative of the phenomenon may by no means be treated as a proof of a statistically founded regularity.

The following sections will present the results of an attempt at analyzing the English, Polish and French corpora of o-words in order to test the claim that o-words containing stops tend to refer to brief and abrupt sounds or actions, whereas onomatopoeic words not containing stops tend to imitate prolonged sounds and actions. Other phonosymbolic means of expressing the referent-sound length will also be examined in relation to the sound-symbolic properties of stops.

3.1.1. The phonosymbolic role of stops in English o-words

When analyzing the corpus of English o-words with respect to the sound-symbolic role performed by stops it must be emphasized that there are also other ways in which the relative length of the referent sound or action is rendered by phonosymbolic means. A sound-symbolically oriented analysis of the 250-item corpus reveals that whenever a word contains a prolonged vowel as in whoosh, a diphthong as in cry, or what may be called a prolonged consonant as in grr (usually marked by repetition of the letter in the graphemic form), it tends to imitate a prolonged sound or action even when it contains a stop/stops at the same time. It thus seems that the symbolic message of abruptness carried by stops tends to be blocked or made ineffective whenever a word containing a plosive also contains one of the above mentioned prolonged segments. The 255-item corpus of English o-words contains 55 such words and they are listed separately in Appendix Two.

Another group of o-words which constitute special cases and should be discussed before the analysis of the phonosymbolic role of stops is performed, is a group of onomatopoeic words (many of which function both as nouns and verbs) ending in a Consonant+l cluster (as in bubble, crackle, rattle, whistle, etc.) and in a C+r cluster as in stutter, jabber and murmur. Malkiel (1990) observed that certain English verbs (he did not indicate that a significant number of them are of an onomatopoeic character), namely those ending in C+l, have much in common with respect to their meaning. As Malkiel phases it, they all refer to states and activities which are "unusually, even inordinately and sometimes perversely, exciting to the average onlooker, speaker or listener" (Malkiel 1990: 184). He also says that these states and activities are "mostly abnormal", "atypical", "occasionally grotesque" or, at least "unlikely to leave the average witness wholly indifferent" (Malkiel 1990: 184). Indeed, all the items ending in C+l which have been isolated from the corpus of English o-words do conform to the above quoted features of actions or states referred to. In fact, in the case of an entirely onomatopoeic corpus they will mostly refer to startling or unusual sounds or actions (e.g., gargle, rattle, gobble, sizzle etc.). Malkiel further suggests that the properties shared by the verbs included in the group discussed above are closely tied to the phonosymbolic function performed by the C+l ending.

The analysis of the English corpus of o-words enabled me to isolate another group of verbs which exhibits many similarities to the one discussed above. It is a group of words ending in C+r. The items included in this group will generally conform to the semantic guidelines which applied for words ending in C+l. The members of this group, however, will even more often imitate complex, repeatedly occuring sounds resulting most often from an unusual manner of speech as in stammer, stutter whisper and chatter and sometimes from other sources as influtter, clatter, sputter and patter. Considering the above indicated features shared by all onomatopoeic words ending in C+r, it seems plausible to suggest that the C+r ending plays the phonosymbolic role motivating the choice of referents for o-words containing this ending just as it was proposed by Malkiel for the verbs with the C+l ending. It is very relevant for the subsequent discussion of the phonosymbolic function carried by stops that both o-words ending in C+l a nd those ending in C+r always refer to actions which produce complex, prolonged sounds and this phenomenon is never affected by the presence of stops. This suggests that the phonosymbolic function performed by liquids in the C+{l/r} ending takes precedence over the phonosymbolic function of brevity otherwise carried by stops.

There are seven cases of o-words in the English corpus, which do not contain stops but contain affricates: chuff, jar, smooch, munch, choo choo, chirr and achoo. It is assumed here that affricates are, in a way, an intermediate stage between stops and fricatives from the point of view of the way in which the air is released by the articulators in the oral cavity. That is why affricates are, on the one hand, unable to play consistently the sound-symbolic role assigned to plosives and, on the other hand, their presence in o-words not containing plosives may disturb these words' general tendency to refer to prolonged sounds. That is why they will not be included in the list of o-words containing stops as well as in the list of words not containing stops prepared for the assessment of the phonosymbolic relevance of the presence/absence of stops in onomatopoeia, in order not to bias the results of that analysis.

To conclude the above discussion, it has been shown that there definitely are cases in which the hypothesis granting stops an unrestricted sound-symbolic function of indicating shortness and abruptness of sounds imitated in onomatopoeia will not be valid. It is then postulated that stops appearing in English onomatopoeic words will indicate the shortness and abruptness of sounds referred to, provided they do not appear in words containing long vowels, diphthongs and prolonged consonants. Also, the presence of C+l and C+r suffixes will pose a restriction to the phonosymbolic function performed by stops. Considering the above restrictions only the o-words that do not violate them were included in the corpus on which the phonosymbolic role performed by stops was to be tested. These items were further divided into two groups: one of them contained words in which stops were present and the other one was made up from the remaining o-words. Each of the two groups was then divided into two subgroups, or categories; t he first one contained o-words referring to short and abrupt sounds and actions and the other one was made up from words which referred to prolonged sounds or actions. The items in each group were then counted. The results of the analysis are presented in Table 1 below.

Table 1

The phonosymbolic role of stops in English o-words.

                                Onomatopoeia              Onomatopoeia
                              containing stops           not containing
                                                        stops
                                N         %              N        %

Number of o-words               92      100.0            46     100.0
o-words referring to            81       88.0             3       6.5
short sounds and actions
o-words referring to            11       12.0            43      93.5
prolonged sounds
or actions

As seen in Table 1 there is a very significant preference for o-words containing stops to imitate short and abrupt sounds and actions. O-words not containing stops, on the other hand, in most cases tend to imitate prolonged sounds. Interestingly, the majority of o-words not containing stops and referring to prolonged sounds (74%) contain the same types of prolonged segments, which were said to bar stops from performing their phonosymbolic function.

3.2. The phonosymbolic role of stops in Polish o-words

The analysis of the sound-symbolic function of stops in Polish o-words was carried out in the same fashion as the analysis performed for English o-words. First, as it was the case with the English corpus, the presence of a prolonged vowel or consonant would normally block the notion of brevity and abruptness carried by stops and it would indicate a prolonged referent sound regardless of the presence of stops. In the Polish corpus of onomatopoeia, however, there are only five such cases, namely bu(u), pi(i), be(e), brr, and tss, with all the three instances of the prolonged vowel being optionally realized in the graphemic form.

Also, as was the case with English onomatopoeia, those items which did not contain stops but contained affricates were, due to their intermediate character, excluded from the analysis of the phonosymbolic relevance of the presence/absence of stops in Polish o-words. I did not, however, exclude those cases of o-words whose only affricate was the sound [t] being part of the infinitival suffixes as in {-ec}, {-oc} or {-ic}. I decided not to do so in view of the fact that the purely grammatical function performed by these suffixes would be superior to any possible symbolic connotations of affricates as they might be exhibited should these sounds appear without any grammatical motivation.

A phonosymbolically oriented search of the corpus allowed me to identify a group of o-words which share the {-ot} suffix and in terms of their meaning have much in common with the English o-words ending in C+l and C+r, which were discussed in the previous section. They also refer solely to prolonged and complex sounds and actions as well as they often leave the listener startled or uneasy about their nature as in belkot, gruchot, grzmot, lomot. Although most of the Polish o-words ending in {-ot} are nouns, the majority of them may be rendered as verbs when equipped with the {-ac} suffix, e.g., furkot-furkotac, lomot-lomotac, which makes them even more similar to the English o-words ending in C+r and C+l, which in most cases function as both nouns and verbs. Two of the words included in this group (there are 25 altogether, all of them listed in Appendix Three), namely plegotac and mamrotac are different from the others in that they are already in a verbal form which may not be changed into a nominal form by ex tracting the {-ac} ending but they still clearly belong to the rest of the group. The similarity between the Polish o-words ending in {-ot} and the English ones ending in C+l and C+r is made even more pronounced by the fact that a number of the former translate directly into the latter, e.g., furkot-flutter, mamrotac-mumble, chichotac-chuckle, giggle etc. Summing up, it seems plausible to claim that the {-ot} suffix does have a symbolic effect on the Polish o-words it appears in.

What is pertinent from the above discussion of Polish o-words ending in {-ot} to the analysis of the phonosymbolic function of stops appearing in the same words, is that as it was the case with the English o-words ending in C+l and C+r, they all refer to complex and prolonged sounds and actions with the stops having no apparent phonosymbolic effect in their presence. The hypothesis assigning stops the phonosymbolic function of carrying the notion of brevity and abruptness will then again have to be restricted only to the cases when other, more effective phonosymbolic means of indicating the length of sounds imitated in o-words are not present. The following Table 2 presents the results obtained for the phonosymbolic function of stops in Polish onomatopoeia. The division into groups and categories is the same as it was done for the English o-words in Table 1.

Table 2

The phonosymbolic value of stops in Polish o-words.

                           Onomatopoeia              Onomatopoeia not
                          containingstops               containing
                                                     stops
                            N         %             N           %

Number of o-words          132      100.0           32        100.0
o-words referring to        96       72.7            7         21.9
short sounds and
actions
o-words referring to        36       27.3           25         71.1
prolonged sounds
or actions

It is clearly visible that although there is a tendency for Polish o-words containing stops to imitate short and abrupt sounds, it is much less pronounced than it was for the equivalent class of English o-words. This may be at least partly explained by the fact that 14 out of the 36 o-words containing stops and referring to prolonged sounds and actions come from the hunter's jargon (cf. Kwiatkowski 1992) and imitate usually phonetically complex voices of birds and other animals. Most of these words contain stops, which repeated several times throughout the word, help reinforce the rhythmic complexity of the referent sound. It seems that the sound-symbolic properties of this group of o-words fall to a large extent within the model of structural symbolism, where the rhythmic patterns play a significant role in conveying the symbolic message (for a discussion of structural and elemental/segment-oriented sound symbolism cf. Taylor -- Taylor 1965). In structural symbolism also the length of the sounds imitated ten ds to be rendered by the variation of word length rather than various sound qualities. Among the 96 Polish o-words containing stops and imitating short sounds, 75 (78%) are monosyllabic and only 2 words (2%) contain three syllables (these two words imitate a series of distinct, short sounds and that is why they were included in that list). On the other hand, only 5 words (13.9%) out of the 36 items containing stops and referring to prolonged sounds are monosyllabic and as many as 8 (22.2%) contain three syllables. This clearly indicates that in the case of Polish o-words, word length is another factor which may take precedence over the phonosymbolic function of abruptness performed by stops.

3.3. The phonosymbolic role of stops in French o-words

The analysis of the phonosymbolic role of stops in French onomatopoeia was conducted along the same lines as the two previously presented analyses performed for English and Polish onomatopoeia.

It was observed that also in French o-words, the use of prolonged vowels and/or consonants would indicate the imitation of a prolonged sound or action.

There are 21 such words in the French corpus, which at the same time contain stops whose phonosymbolic function of indicating shortness and abruptness is suspended in the presence of prolonged segments.

In the case of French onomatopoeia no groups of o-words were identified to possess sounds or morphemes which would make stops phonosymbolically ineffective in a shared lexical context.

It is word length, however, which as a means of indicating the length of sounds imitated seems to be much more often employed in French onomatopoeia than in English or even Polish. The repetition of the same or slightly altered syllable two or three times, for example, is used here as a means of indicating that a particular referent is to be identified as a series of clearly separate short sounds (hence quite a few bisyllabic and three syllable words containing stops in my "short sound" category). Words with more than three syllables, however, (there are no such words among Polish and English onomatopoeia) always refer to relatively long sounds, no matter if they contain stops, or not. In these words, stops will function mostly on a structurally-symbolic plain, mainly helping to imitate the rhythmical pattern of the referent as in taratata (the sound of a trumpet), or patapatapon (imitation of a horse's galloping). Apparently in such polysyllabic words, elemental sound symbolism as exhibited in the phonosymbo lic message of shortness normally carried by stops, gives way to structural symbolism, with the phonosymbolic function of brevity otherwise performed by stops being switched off.

Considering the above, it will be postulated that stops occurring in French o-words will imitate the shortness and abruptness of the referent sounds as long as they do not appear in words containing prolonged segments (vowels and consonants) and/or words containing more than three syllables. Table 3 presents the results obtained for the phonosymbolic function of brevity as exhibited in stops occurring in French o-words which comply with the above restrictions. The division into groups and categories is parallel to that seen in Tables 2 and 3.

Table 3

The phonosymbolic value of stops in French o-words.

                                  Onomatopoeia        Onomatopeia not
                                containing stops      containing stops
                                  N         %             N

Number of o-words                 75      100.0          45
o-words referring to short        66       88.0           5
sounds and actions
o-words referring to               9       12.0          40
prolonged sounds or
actions

                            Onomatopei
                               a not
                            containing
                              stops
                              %

Number of o-words           100.0
o-words referring to short   11.1
sounds and actions
o-words referring to         99.9
prolonged sounds or
actions

The data presented in Table 3 show that also in the case of French onomatopoeia there is a very clear tendency for stops to occur in o-words imitating short sounds and actions. The absence of stops, on the other hand, will in most cases indicate that a given word refers to a prolonged sound or action.

3.4. Conclusions

The three analyses of the phonosymbolic ways of indicating the relative referent-sound length in English, Polish and French onomatopoeia have demonstrated that there is an obvious tendency for o-words in these languages to be phonosymbolically affected by the presence or absence of stops. The former would in most cases indicate that a given word refers to a short and abrupt sound or action, whereas the latter would be typical for o-words referring to prolonged sounds or actions. It has, however, been observed that plosives do not perform their phonosymbolic function of abruptness in all contexts. Whenever a word containing a plosive also contains a prolonged vowel, a diphthong or a long consonant, it refers to a prolonged sound or action. Also, English o-words ending in C+l and C+r, as well as Polish onomatopoeia ending in {-ot}, refer to complex and prolonged sounds despite the fact that most of them contain plosives. Word length (given in syllables) is another factor which in some cases takes precedence ove r stops as a phonosymbolic means of indicating the length of the sound imitated. This is particularly visible in French and Polish o-words, as in the English corpus 97% of all o-words are either mono- or bisyllabic, which means that onomatopoeic words in this language do not favor word length as a phonosymbolic device to the extent in which it is visible in Polish and French, where one can find words words containing three, four or even (in the case of French) five and six syllables. In order to demonstrate how syllabic word length affects the phonosymbolic message relating to the imitated sounds' length in o-words containing plosives and those that do not contain them, the average number of syllables per word was counted for the items within each of the groups and categories from Tables 1, 2, and 3. Table 4 presents the results of this analysis.

Table 4

The relation between the syllabic length of o-words and the length of
the sounds imitated in English, Polish and French onomatopoeia.

                       English o-words           Polish o-words
                   o-words       o-words   o-words     o-words
                   cont.         not       cont.       not
                   stops         cont.     stops       cont.
                                 stops                 stops

The ratio of       1.03          1.0       1.21        1.4
syllables to
words in o-words
ref. to short and
abrupt sounds

The ratio of       1.27          1.1       12.0        1.44
syllables to
words in
o-words ref. to
prolonged sounds

                       French o-words
                   o-words      o-words
                   cont.        not
                   stops        cont.
                                stops

The ratio of       21.59        1.0
syllables to
words in o-words
ref. to short and
abrupt sounds

The ratio of       2.66         1.42
syllables to
words in
o-words ref. to
prolonged sounds

The figures in Table 4 show that o-words containing stops and referring to short sounds and actions are in all three languages shorter in terms of their syllabic length than the words containing plosives and referring to prolonged sounds. The difference, however, is most pronounced in Polish and French onomatopoeia. Interestingly, words not containing plosives and referring to prolonged sounds are, with the exception of French o-words, not significantly longer than those referring to relatively short sounds.

The above shows that syllabic word length as a feature of structural symbolism does constitute a significant phonosymbolic factor which in the case of polysyllabic words tends to deprive stops of their phonosymbolic function of indicating the brevity of the sounds referred to.

The final conclusion to the discussion presented in this paper will be that stops as carriers of the notion of brevity may not be analyzed in isolation from other phonosymbolic devices, which in many cases are able to affect, or even neutralize their sound-symbolic function. One might then postulate a hierarchy of phonosymbolic devices which possess different ranks on what might be called a scale of sound-symbolic effectiveness. It seems plausible that this ranking system will also be exhibited outside the range of the sound-symbolic means of expressing the relative referent sound length in onomatopoeia discussed in this paper and will eventually help solve many puzzles and inconsistencies in the behavior of phonosymbolic devices encountered in the sound symbolism research.

APPENDIX ONE (1)

French onomatopoeia

1.1. O-words containing stops and referring to short and abrupt sounds.


badaboum
badadam
bang
baoum
bing
bip bip
bloum
bouach
boum
cac cac
chut
clac
clap clap
clic
cloc cloc
cou cou
couic
cot
couac
couach
coin coin
crac
cric crac
cri cri
ding ding
flac
fli flap
floc
glou glou
hep
hic
hop
houp
kic kic kic
paf
pan
patapouf
patatrac
patatras
pflac
pif
pif paf
pin pon
plaf
plouf
pouet pouet
pouf
pouh
pou pou pou
prout
prt
pst
tac tac
teuf teuf
tia tia
tic tac
tin
tio tio
tiou tiou
ti ti ti
toc toc
top
tskuo tskuo
tut
yop
youp
1.2. O-words containing stops and referring to prolonged sounds or
actions


co co co
coin coin
cui cui cui
ding dong
piou piou
rataplan
tararan
tireli
tut tut
1.3. O-words not containing stops and referring to prolonged sounds


ah
ahan
aie
chnouff
choo choo
eh
ha ha
he he
hein
heu
hi han
hi hi
ho ho
hou hou hou
hon
ihihih
meuh
miam miam
miaou
mmmmmm
oh
ohe
ouaf
ouah
ouf
ouille
ouin ouin
ron ron
rron
rrou
rrr
snif snif
sssss
vlouf
vraoum
vroum
woo woo
zoum
zzzz
1.4. O-words not containing stops and referring to short and abrupt
sounds.


fla
han
hem
ra
vlan

1.5. O-words in which the phonosymbolic function of brevity carried by stops is suppressed by other phonosymbolic means

1.5.1. O-words containing over 3 syllables


cocorico
patapatapon
patati patata
taratata
taratantara
1.5.2. O-words containing prolonged segments (vowels and consonents)


bah
be bee
beuh
booff
brr
bzz
coa
crr
grr
kss kss
llpp
oooc
oops
pff
pfft
pffrut
pffut
pschhh
pschitt
tss
vroutt
APPENDIX TWO

English onomatopoeia

2.1. O-words contanining stops and referring to short and abrupt sounds


bang
bark
bash
belch
bong
bump
burp
chirk
chirp
chuck
chug
clack
clang
clank
clap
click
clink clank
cluck
crack
crash
croup
crunch
crunk
crush
dash
dit
dot
drip
fart
flap
flip
flop
gasp
glug
gnash
gulp
gush
hack
hop
klunk
knock
lap
pit
plink
plop
pluck
plump
plunk
pull
pop
pow
putt
rap
skim
skip
slap
slurp
smack
snap
snip
spit
splash
splat
stomp
swat
tap
thud
thump
thwack
tick tack
tick tock
tramp
grab
trip
whack
whop
yap
yelp
yip
zap
zip
2.2. O-wards containing stops and referring to prolonged sounds


ding dong
drum
honk
lisp
rant
rataplan
ratatat
scratch
strum
trumpet
twang
2.3. O-words not containing stops and referring to prolonged sounds


arf
fizz
hail
heehaw
hiss
hoarse
howl
huff
low
lull
mew
miaow
moan
moo
neigh
ring
roar
shoo
sigh
sing
sizzle
sneeze
sniff
snore
sooey
sough
swish
swoosh
vroom
wail
waul
whang
wheeze
whiff
whine
whinny
whirr
whiz
whoosh
woof woof
yawn
yell
zz
2.4. O-words not containing stops and referring to short and abrupt
sounds


hem
shush
slash
2.5. O-words containing affricates (not containing stops)


achoo
chirr
chuff
jar
choochoo
munch
smooch
2.6. O-words in which the phonosymbolic function of brevity carried by
stops is suppressed by other phonosymbolic means

2.6.1. O-words containing long segments


baa
bay
beep
bellow
blare
bleat
blow
boo
boohoo
boom
bowwow
bray
buzz
bzz
caw
cheep
claw
clear
cloop
coo
creak
croak
cry
cuckoo
drone
eek
gnarr
grind
groan
growl
grr
hoot
paw
peal
peep
pipe
pow
prate
pule
purr
scrape
scream
screech
shriek
squeak
squeal
teehee
toot
troat
tweet
ululate
whoop
2.6.2. O-words ending in C+l.


babble
bubble
burble
cackle
chuckle
crackle
dandle
dribble
drivel
gabble
gargle
giggle
gnarl
gobble
grumble
gurgle
jingle
mumble
popple
pumble
prattle
rattle
rustle
sizzle
snarl
sniffle
snuffle
tinckle
trickle
warble
whistle
2.6.3. O-words ending in C+r.


chatter
clamor
clatter
flutter
gibber
jabber
mutter
murmur
nicker
patter
pitter
snicker
spatter
sputter
stammer
stutter
titter
twitter
whimper
whisper
APPENDIX THREE

Polish onomatopoeia

3.1. O-words containing stops and referring  to short and abrupt sounds


bach
bek
bec
bek
brzdek
brzek
buch
bul
bum
bums
bzdzic
bzik
chap
chlap
chlast
chlip
chlup
chlust
chrochtac
chrostac
chrup
chrzaknac
cmok
cyk
cyt
czknac
czlap
fuk
gda
gruch
gul
hop
huk
kap
klak
klap
klaskac
klik
klop
klump
kolnac
kolatac
kop
krach
krop
krzaknac
ku ku
kwa
kwo
lkac
lub
lup
op
pac
pach
pac
palnac
pam
parsknac
pek
pif paf
plask
plop
plum
plusk
pok pok
prask
pruk
prztyk
pst
psyk
puch
pum
pyk
pyr
pyrk
rabnac
siaknac
siekac
skrzek
smagac
smak
stek
strzal
stuk
szast prast
szczek
szczkac
tchnac
tfu
tik tak
trach
tryk
trzask
tst
tup
3.2. O-words containing stops and referring to prolonged sounds


brukac
buczec
burczec
chrapac
chrypiec
ciurkac
ciurlikac
dachtac
dac
dedac
gorlic
gwar
jodlowac
krakac
krekorac
krerac
krzyk
ksykac
kwilic
puchac
rzepolic
seplenic
siorbac
skomlec
skrobac
skwierczec
smarkac
syk
szelest
swirkac
terlikac
tokowac
trelowac
turkac
wrzask
zlobic
3.3. O-words not containing stops and referring to prolonged sounds


chi chi
chichrac
cho cho
chuchac
frr
fur
me(ee)
miau
mu(uu)
musowac
rumor
rurac
ryczec
rzezic
rzec
ssac
szloch
szmer
szorowac
szum
szur
wiac
wrr
wyc
ziac
3.4. O-words not containing stops and referring to short and abrupt
sounds


chrum
hau
hm
rum
rym
sza
zachlysnac
3.5. O-words containing affricates (not containing stops)


ciach
czochrac
cwir
dzwonic
dzyn
mruczec
ryczec
trzec

3.6. O-words in which the phonosymbolic function of brevity carried by stops is suppressed by other phonosymbolic means

3.6.1. O-words containing long segments


bee
ba(a)
brr
bu(u)
pi
tss
3.6.2. O-words ending in -ot.


belkot
brzechot
bulgot
chichot
chrobot
klekot
druzgot
furkot
gruchot
grzechot
grzmot
jazgot
lomot

(1.) The notion of brevity is realized for single instances of a given sound whenever there is an option of a repeatedly occurring sound. Some cases may seem controversial with respect to the choice of category, mostly due to the relative character of the categories of short sounds and long sounds.

REFERENCES

Brown, Roger

1958 Words and things. New York: The Free Press.

Jespersen, Otto 

1933 Linguistica. Selected papers in English, French and German. Copenhagen: Levin and Munksgaard.

Kwiatkowski, Pawel

1992 The phonology of onomatopoeia: A Polish-English contrastive study. [Unpublished M.A. thesis, Nicholas Copernicus University, Torun.]

Leech, Geoffrey H.

1969 A linguistic guide to English poetry. London: Longman.

Malkiel, Yakov

1990 Diachronic problems in phonosymbolism. Philadelphia: Benjamins.

Sapir, Edward 

1929 "A study in phonetic symbolism", Journal of Experimental Psychology 12: 225-239.

Sobkowiak, Wlodzimierz

1990 "On phonostatistics of English onomatopoeia", Studia Anglica Posnaniensia 23: 15-30.

Taylor, Insup K. - M. Taylor

1965 "Another look at phonetic symbolism", Psychological Bulletin 64: 413-427.

Zuchowski, Rafal

1993 The phonology and selected aspects of sound symbolism in English, Polish and French onomatopoeia. [Unpublished M.A. thesis, Adam Mickiewicz University, Poznan.]