Interjection
1. A general overview of interjections
Interjections (in Polish wykrzykniki, literally ‘exclamations’) form a heterogeneous class in terms of morphology, syntax, and semantics. Their grammatical descriptions focus on two criteria – the fact that interjections are not inflected, and their syntactic independence viewed as their capacity to function as independent utterances not conditioned by their verbal context. One common denominator for the majority of semantic descriptions of interjections is their connection to the emotional and volitional spheres and the search for the iconic motivation for their meaning in the elements of their forms. Further discussion of this class in this chapter disregards any views assuming that interjections do not belong with the linguistic system (cf. e.g. Wajszczuk 2010: 24).
2. Interjections in the pre-war grammars of the Polish language
The basic classification of interjections which left its mark on both the grammars of the Polish language from the beginning of the 20th century (Krasnowolski 1909: 153; Kryński 1910: 349; Szober 1923: 160; Łoś 1925: 137; Gaert ner 1938: 203), and the most recent descriptions of this part of speech (Strutyński 1997: 143; Grochowski 1988: 86, 1992, 1993, 1997: 14), assumes that this group of expressions consists of two basic classes: primary interjections (interjections ‘proper’, not derived from other expressions) and secondary interjections (historically traceable to other units). This division, proposed by the German psychologist and philosopher Wilhelm Wundt (Wundt 1904: 307–309; quoted in Bednarczyk 2014: 83–84), is based on morphological criteria. It assumes that the first of the two categories outlined above contains interjections that are not historically motivated (e.g. ach!, oj!), while the second group is made of interjections derived from expressions that belong to other grammatical classes (e.g. rety!, rany!, Matko Boska!). However, the criteria proposed in more recent literature to define interjections proper differ from those laid down by Wundt. This is the case, for instance, of Grochowski’s syntactic classifications, where the group of interjections referred to by the author as interjections proper is also inclusive of expressions with a clear etymology, such as Matko Boska! or holender! (Grochowski 1992: 157).
The solution offered by Wundt is not sufficient to fully account for the properties of Polish interjections which form an internally diversified class. This is evidenced, for instance, by the fact that the oldest grammars of Polish where such a division is found typically distinguish between more than two classes of interjections. The descriptions and classifications of interjections, both in pre- and post-war studies, were based on diverse and imprecise criteria. When characterising this group of expressions, the authors often referred to a plethora of functions that interjections may perform in utterances (e.g. expressing emotions, expressing one’s intent, mimicking sounds). What is more, most studies contain more or less detailed reflections on inflection and syntax – even the oldest grammars contain information that interjections are, among other things, uninflected (Kryński 1910: 348; Szober 1923: 160), can function as independent utterances or even have the quality of full sentences (Krasnowolski 1909: 153) and do not enter into any syntactic relations with other expressions (Kryński 1910: 348).
In his “Systematyczna składnia języka polskiego,” Antoni Krasnowolski (Krasnowolski 1909: 153) distinguishes between the three classes of interjections to which he assigns the meaning of specific types of sentences. These are a) interjections proper, which “express an affection or, in other words, a feeling,” and have the meaning of an exclamation (e.g. ach!, o!,
ha!), b) sound gestures that are often accompanied by a gesture and which have the meaning of an exclamatory sentence (e.g. cyt!, hej!, hola!, nuże!, naści!, marsz!, wara!, fora!) and c) mimetic sounds, which “have the meaning of an indicative sentence, since they evoke the idea of the subject who provoked such a sound and the action which triggered that sound” (e.g. trzask, plusk, bęc, miau, be, me).
Similar distinctions may be found in Adam Antoni Kryński’s “Gramatyka języka polskiego” (Kryński 1910: 348), where the author divided interjections into interjections that
a) express feelings (e.g. o, och, a, ach, ha, ho, aj, oj, niestety, biada, przebóg, brawo, e, fe, pfe, tfu),
b) manifest an intent (e.g. e, ej, hej, hejże, hola, no, nuże, marsz, na, naści, wara, fora, precz, won, huzia, basta, cyt, psyt, sza, wio, kse, hetta, prr, ani mrumru), and c) mimic the sounds of nature (e.g. bęc, brzdęk, paf, fajt, puf, pstręk, chlast, szast, brdum, miał, be, me, kuku, ku-
kuryku, dyń dyń, hm hm, patataj, miau miau). Unlike Krasnowolski, Kryński does not identify interjections proper with interjections that express feelings, but rather treats them as a superior category – “interjections are mostly mechanical articulations, such as o, och, ha,
ej, hej…… and these may indeed be called interjections proper; some of the interjections of feeling and intent, however, represent fossilized forms of other parts of speech, e.g. przebóg, biada, brawo […]” (Kryński 1910: 349).
Stanisław Szober’s “Gramatyka języka polskiego” of 1923 only contains perfunctory information on the significance of this part of speech – “Words that are signs for affective states are called interjections, e.g. o! ach! aj! e! a!” (Szober 1923: 98). At the same time, somewhat further on, in a section entitled the “Science of word formation (Nauka o budowie wyrazów),” the author writes that – as invariable expressions deprived of inflection and “word-building forms,” interjections may practically be identified only on the basis of semantic criteria – “interjections, in substantially the same way as prepositions and conjunctions, do not have any formal properties, they are words without form; this is why we identify them only on the basis of the meaning specific to them” (Szober 1923: 160). The author describes the division into interjections proper and secondary interjections in relatively high detail, taking account of the formal-semantic dependencies characterising the pathway typical of many interjections – the shift from other part of speech classes.
Derivative interjections, owing to their connection to other parts of speech, manifest certain formal properties externally, but are deprived of their proper meaning. This is why the category of interjections creates homogeneous combinations of phones which internally, that is semantically, cannot be divided into smaller units, and as such are also deprived of any explicit morphological structure, as the morphological structure of a word is a combination of the semantic units specific to it; once such units loose the semantic content they once possessed, even though externally they continue to be distinguishable by the linguistic awareness, they no longer have the initial nature of their morphological constituents (Szober 1923: 160).
In his “Gramatyka polska,” Jan Łoś (Łoś 1925: 137) found Wundt’s classification to be superior, while pointing out that interjections proper do not form a homogeneous class –
An interjection proper is the voice emitted involuntarily by a person under the influence of affect. In literary language, the so-called interjections proper, to some extent normalised or stylised, stand closest to such voices: they include a, ach, och, ha, hej, hm, fe, brr, etc. They have no etymological connection with the elements of names or verbs. This category is also inclusive of the interjections that accompany deictic gestures: o!.
On the other hand, when it comes to the class of conventionalised interjections, the author sees their function as merely limited to that of communicating feelings, which is what he identifies with etymological motivation for this class of interjections. –
Words that have the function of interjections include also those which – apart from conceptual content – have or used to have affective properties and, over time, became conventional
signs for certain feelings, such as. biada, niestety […], przebóg, masz tobie, precz, wierę etc.
Apart from these two basic categories, Łoś distinguishes an additional one, consisting of onomatopoeic interjections – “The so-called onomatopoeic interjections, allegedly imitating sounds emitted when performing certain activities, form a separate group […].”
The section dedicated to interjections in Henryk Gaertn er’s “Gramatyka współczesnego języka polskiego” (Gaert ner 1938: 203) lacks a systematic classification of this part of speech; the author included, however, some rather original general comments. Apart from concluding that the basic function of interjections is to express the emotional states of the speaker and reporting the division into primary and secondary interjections based on the existence of semantic stems attributable to the latter, Gaert ner highlights the diversified nature of this class from the perspective of its “scope”.
The scope of interjections represents the set of affective states to which interjections refer. And thus, for instance, the scope of fe represents the various states of disgust. The scope of some interjections is very vast, and thus, e.g. an interjection may refer to a variety of states of joy as well as sadness, which are its referents.
Yet another interesting thought presented by the scholar regards the distinction of a subcategory of musical exclamations, which “are conditioned by a certain emotional state, but, at the same time, especially in the case of humming, are primarily musical in nature, and
thus constitute an articulatory background to a certain melody or rhythm, e.g. dana, hopsa dana, dylu, dylu etc.”
3. Interjections as a part of speech; syntactic classifications
In the post-war and most recent linguistic literature, the grammars of most relevance to interjections have been those classifying parts of speech on the basis of homogeneous and coherent criteria (mostly syntactic ones), as well as studies dedicated to interjections themselves that built upon the assumptions made in such grammars. The most detailed description of this class of expressions in the Polish language made on the basis of syntactic criteria may be found in the studies by Maciej Grochowski (Grochowski 1986, 1987, 1988, 1992, 1993, 1997).
Grochowski defines interjections as invariable lexemes that function as separate utterances, independent from the verbal context (Grochowski 1992: 156, 1988: 87, 1997: 14). The fact that interjections may be used independently differentiates them from all of the remaining classes of invariable expressions apart from appositions (in Polish: dopowiedzenie);
what makes them stand apart from the latter is their contextual independence (Grochowski 1988: 87). The author distinguishes between four subclasses of interjections: primary interjections, or interjections proper (e.g. ach, ej, fuj, uf), parenthetic interjections (e.g. cholera, kurwa, psiakrew, rany boskie), onomatopoeic interjections (e.g. bzz, bęc, chaps, łubu-du, pac) and appellative interjections (e.g. huzia, jazda, precz, won, wara). The expressions from the first two of the abovementioned classes may only occur as independent utterances, while onomatopoeic and appellative interjections may occur in multi-element utterances as their constitutive parts (e.g. Królik kic po trawie., Wszystkie książki łubu-du na podłogę., Huzia na
nich!) (Grochowski 1993; 1997: 14). One characteristic feature of parenthetic interjections
(which differentiates them from interjections proper) is their capacity to occur between any autosyntagmatic components of an utterance (independently of the order, such interjections do not form any syntactic relations with any of the constituents of the utterance) (Grochowski 1988: 90; 1992: 158; 1993: 299).
This division does not fully align with the previous versions of Grochowski’s classifications of interjections (differences concern the terminology adopted by the author, as well as – to a certain extent – the classification criteria). Previously, this researcher (Grochowski 1988) dichotomously divided this class of parts of speech into interjections that a) cannot co-occur
with other expressions in the same utterance; and those that b) allow for such co-occurrence.
Within the first of these classes, the scholar distinguished between a1) appeals – interjections that may bind expressions in vocative (e.g. Dzień dobry, panie dziekanie, Pa, synku!, Serwus, bracie!) and a2) parenthetical interjections which do not allow for such combinations (e.g.
*Holender, Piotrze!, *Kurczę, człowieku!, *Psiakrew, dziewczyno!). The latter group was further subdivided into b1) onomatopoeias – lexemes capable of constituting indicative utterances (e.g. bęc, czmych, cf. Piotruś bęc na dywan., Zając czmych do lasu.) and b2) predicative interjections (also referred to by MG as volitional interjections (Grochowski 1992: 156) – lexemes that cannot constitute indicative utterances (e.g. a kysz, dosyć, huzia, jazda, precz, wara, won, wynocha) (Grochowski 1986: 40–44, 1988).
Grochowski’s definition of interjections is based on the two descriptions of this class of expressions, resulting from the functional classifications of Polish lexemes. The first one, by Roman Laskowski, assumes that interjections are asyntagmatic lexemes (expressions that primarily function as independent utterances, and do not enter into any syntactic relations
with other elements of a text, e.g. hej, bęc, halo, bzz), while also being non-contextual, i.e. they do not imply the existence of a text and may independently form a complete text themselves, e.g. uwaga, psst, brzęk (Laskowski 1984: 30–31). Laskowski’s classification distinguishes between two subclasses of interjections. The first consists of appeals which may secondarily be combined with imperative forms (e.g. Halo, poczekajcie!) and with the vocative (Uwaga, ojcze!), while the other is made of onomatopoeias that may secondarily occur in the function of a sentence-forming component (e.g. Dziecko bęc jak długie, Kula tylko bzz koło ucha)
(Laskowski 1984: 31). The second of the abovementioned classifications was developed by Zygmunt Saloni and Marek Świdziński, who defined interjections as a “class of invariable lexemes whose only form is used independently” (Saloni, Świdziński 1985: 95). The scholars further added that interjections (apart from the vocative forms of nouns and some forms of certain verbs) “have empty connotation,” meaning that “they do not announce any syntactic units – neither as dependent, main nor coordinate constituents” (Saloni, Świdziński 1985: 216).
There are some doubts concerning the asyntagmatic property of interjections when it comes to onomatopoeias as well as appellatives that may perform the predicative function in a sentence. In Polish linguistic literature, onomatopoeias have been particularly extensively studied (cf. e.g. Pszczołowska 1975; Siatkowska 1976, 1977; Rittel 1986; Grochowski 1996; Bańko 2008, 2009). Many researchers, who argued their stance relying mostly on syntactic and semantic argumentation, have proposed that this subclass be distinguished from the class of interjections (Siatkowska 1976, 1977, 1985; Bednarczyk 2014: 88); some grammarians have described these expressions as a particular subclass of verbs (e.g. Klemensiewicz 1937: 106; Jodłowski 1971: 88, 1976: 79). In the literature on the subject, we may also come across claims that the syntactic properties of onomatopoeia and onomatopoeic verbs derived from them are identical (e.g. ciach – ciachnąć, puk-puk – pukać) (Jodłowski 1976: 32; Bartmiński 1978: 162, 1981: 8; Boniecka 1977), as well as arguments on the similarity of the syntactic patterns of onomatopoeia, verbs of movement and verbs of speech (Bańko 2008: 103). According to Grochowski, the word-formation dependency is insufficient to conclude that the
valency requirements of expressions that make up such pairs is identical (Grochowski 1998: 93–94). The problem of valency also applies to appellative interjections which, in such sentences as Wara ci od Marii! or Won z pokoju!, impose syntactic restrictions on the nominal forms in their direct or further context. Despite the specific properties of onomatopoeic and appellative interjections, Grochowski rejects the notion that they might be included in the class of verbs, as they cannot be assigned any values of the grammatical categories inherent to verbs, in sharp contrast even to the verbs with a limited inflectional paradigm (such as można, trzeba, szkoda etc.). The researcher does, however, put forward two other solutions that allow for the specific valency-related properties of appellatives in their syntactic classification to be accounted for. The first involves recognizing their capacity to perform the predicative function as a secondary property, while the other consists in forming a separate class of non-verbal predicatives that would include homonyms of appellative interjections (such as huzia, precz, wara) (Grochowski 1992: 157, 1997: 15). On the other hand, to solve the problem of the predicative function assumed by certain interjections, while distinguishing them at the same time from such verbs as trzeba or można, Mirosław Bańko suggests a modification to the definition adopted by Grochowski: “interjections are invariable and utterance-forming lexemes, capable of constituting an utterance either independently or with components dependent on them” (Bańko 2008: 41).
4. Semantic classifications and the semantics of interjections
Anna Wierzbicka was the first to put forward a classification of interjections on the basis of semantic criteria (Wierzbicka 1991: 290–326). The author defined this class in the following way:
An interjection can be defined as a linguistic sign 1. which can be used on its own; 2. which expresses a specifiable meaning; 3. which does not include other signs (with a specifiable meaning); 4. which is not homophonous with another lexical item that would be perceived as semantically related to it; and 5. which refers to the speaker’s current mental state or mental act (for example I feel … , I want … , I think … , I know … ) (Wierzbicka 1991: 290).
What merits our particular attention is item 4 of the foregoing definition which eliminates such expressions as Good Lord!, Good heavens!, Christ!, Hell! or Polish Boże!, Jezus Maria! or Cholera! from the group of interjections. Another problematic group consists of onomatopoeic expressions, which may be used in Polish sentences as a substitute for predicates, interpreted by Wierzbicka as the homophones of interjections which are not interjections in themselves. The author argues that this issue could be solved by rephrasing the problematic item: “(4) which is not homophonous with another lexical item whose meaning would be included in its own meaning (that is, in the meaning of the putative interjection).” In consequence, the inclusion of exclamations such as Cholera! into the group of interjections depends on their semantic explications.
The author distinguished between three basic categories of interjections: 1. emotive interjections, whose main semantic component is ‘I feel something’ and which do not contain the ‘I want something’ component, 2. volitive interjections, based on the ‘I want something’ component, but not having the ‘I feel something component’, 3. cognitive interjections, whose structures contain the component ‘I think something’ or ‘I know something’,
while not containing any emotive or volitive components; many expressions classified into this class have their homophonic equivalents classified as emotive interjections.
Emotive interjections are divided into two subclasses: interjections expressing disgust and similar feelings, such as the Polish fu, fe, tfu, the English yuk or the German pfui, as
well as general purpose interjections represented e.g. in Polish by oj, ojej, och, ach. The author also subdivided volitive interjections into two classes – those directed at animals, e.g. cip-cip,
kici-kici, huzia, wio, prr, sio, and interjections directed at people. The scholar describes the subclass of volitive interjections directed at people by distinguishing further subgroups rep-
resented by the following labels: ‘I want silence’ (sza, pst, cii), ‘I don’t want you in this place’ (won, precz, sio, wara), ‘I want you to jump’ (hop, hopla), ‘urging’ (nuże, hej, hejże), ‘communication over distance’ (hop hop, hallo, ahoj), ‘I give it to you’ (na). Wierzbicka does not propose any further subdivisions in the class of cognitive interjections, including here, for instance, the Polish aha, oho, o, the Russian aga or the English oh-oh.
Yet another classification of interjections based on semantic criteria was developed by Elżbieta Orwińska-Ruziczka (Orwińska-Ruziczka 1992). In this classification, interjections are divided into three main groups a) impulsive, b) imperative, and c) representative interjections (onomatopoeia). Impulsive interjections are further subdivided into sensorial – of almost purely physiological nature (e.g. aj, au, brr), emotive – related to the psychological sphere and expressing mostly emotions (e.g. haha, ech, phi) and intellective interjections – being an expression of the involvement of the intellect (rather than emotions) of the speaker (e.g. hm, uhum). When it comes to imperative interjections expressing the intent of the speaker, the author distinguishes between attention-drawing or demonstrative interjections o, hej-hej, halo, hola), imperative interjections (e.g. pst, stop, jazda) and salutations (e.g. cześć, pa, z Bogiem), subdividing that last category – based on the degree of iconicity – into representative primary and secondary interjections. The former are then divided into conventional interjections (e.g. tik-tak, wrr) and single-word non-conventional forms, while the latter, concerning mostly visual, and sometimes visual-auditory impressions, into sound symbols representing only a type of action without an acoustic element (e.g. mig, chap) and sound symbols representing a type of action and a sound quality as components closely connected to each other (e.g. fru, szast) as well as sound symbols representing metaphorical meanings (e.g. ty
kukuryku meaning ‘you rooster’).
The classification proposed by Monika Krzempek may also be considered as somewhat semantic. Given the heterogeneous nature of the class of interjections, extended by the researcher by adding appositions and verbs not inflected for person (Krzempek 2018: 120), the scholar proposed a description relying on the assumptions of cognitive methodology (with a special focus on prototype theory). In this approach, interjections may be divided according
to the domains in which they are conceptualized. These are the domains of feelings, sound, movement, intent, and scripts. As regards the domain of feelings, the author refers to the six basic emotions proposed by Ekman and Friesen (Ekman, Friesen 1971) identified with the domains where interjections may be conceptualized: enjoyment (e.g. brawo, hura), disgust
(e.g. fuj, fe), surprise (e.g. jejku, rety), anger (e.g. holender, kurde), fear (e.g. Boże, Chryste), and sadness (e.g. ach, au, ojoj). Interjections conceptualized in the domain of sound are di-
vided by Krzempek into three groups – sounds emitted by animals (e.g. kukuryku, me, miau), sounds emitted by humans (e.g. a psik, ble, chlip, cmok), and sounds emitted by instruments, objects and devices (e.g. drr, dzyń, pif-paf). The groups of interjections conceptualized in the domain of movement (e.g. czmych, fik, frr, huzia, hyc, jazda) and in both domains (of sound and movement) at the same time (e.g. bęc, brzdęk, chlust) are not subdivided further in any way. By contrast, Krzempek divided the interjections conceptualized in the domain of intent and used to regulate interpersonal reactions into three categories: interjections used to draw attention to something or to establish contact (e.g. ej, ejże, hej, halo), interjections used in communication with animals (e.g. a kysz, kici, wio), interjections aimed at eliciting a specific action, including: movement in a specific direction (e.g. chodu, huzia, jazda, precz, wynocha), ending a conversation or an action (e.g. basta, hola, kwita, stop), urging somebody, encouraging them to hurry up (e.g. dalej, nuże), or silencing someone (e.g. cyt, pst, sza). Interjections conceptualized in the domain of script, construed as knowledge of specific behavioural patterns, mostly include interjections used in the performance of communicative models (e.g. cześć, halo, pa, aha, no, uhm) (Krzempek 2014; 2018: 126–128).
The discrepancies between the views of linguists as regards the semantics of interjections in a broad sense are not only manifested by the attempts at the internal organisation of this class based on the semantic criteria, but also by the answers to the question of whether interjections carry a meaning in themselves. Apart from the approaches that classify all interjections as meaning-bearing (e.g. Wierzbicka 1991), there are also opinions to the contrary,
with some scholars perceiving the class as devoid of meaning (e.g. Mirowicz 1967: 78; Dąmbska 1975: 77) and some treating interjections as a heterogeneous class from this perspective (Grochowski 1992; Śledź 2004: 265). More or less detailed semantic reflections may be found in the majority of linguistic analyses of interjections, but when it comes to the research on the meaning of this class of expressions, the studies that merit particular attention are those that attempt an explication of the meaning of specific lexical units. The semantic analyses by A. Wierzbicka (Wierzbicka 1991, 1992, 1996, cf. also semantic comments on interjections in Wierzbicka 1969, 1986, 1987) and M. Grochowski (Grochowski 1988, 1992, 1993) are of particular relevance in this context.
Grochowski argues that the expressions classified by him as onomatopoeic, volitional and appellative interjections are not devoid of meaning; to justify this view he proposes semantic representations of several onomatopoeic interjections (fik, frr, chlup, pac) and tens of interjections described by him as volitional (wynocha, won, wara, dość, dosyć, sza, pst, tss, cii, ani mru-mru, bis, hop, hejże, nuże, hopla, hops, hopsa, na, huzia, ej, hej, te, hop, hop, halo, hola, precz, wio, wiśta, hejta, prr) (Grochowski 1993). A semantic analysis of specific expressions from the latter group was preceded by a detailed semantic classification based on the dichotomous divisions of interjections into the subclasses classified by Wierzbicka as volitional interjections directed at people. The first level of this classification assumes a division into 1. interjections containing a semantic component ‘I want to force you to do something’ and 2. interjections which do not contain such a component, further subdivided into 2.1 in-
terjections containing the component ‘I want you to know that…’ (e.g. ej, te, hej, halo1, halo2, hop, hop) and 2.2. those without such a component (hola, precz2). The first of the abovementioned classes is divided into 1.1 interjections implying the concept of negation and 1.2 not implying the concept of negation (e.g. bis, hejże, hopla, hops, hopsa, huzia, na, nuże). The author divides subclass 1.1. into interjections; 1.1.1 comprising a locative component, and further splits this group into interjections that; 1.1.1.1 contain a component ‘I feel something bad towards you’ (precz1, won) and 1.1.1.2. interjections without this component (wynocha, a kysz, a sio, fora) as well as 1.1.2. interjections without any locative component that are also subdivided into 1.1.2.1 containing an ‘I feel something bad towards you’ component (wara) and 1.1.2.2 interjections without this component (e.g. dość, dosyć, pst, tss, cii, sza, ani mru–mru).
Grochowski describes interjections proper as a semantically heterogeneous class (Grochowski 1992: 162). The author developed a list of 33 interjections proper, which he considers as semantically empty expressions (a, ach, aj, au, e, ech, ej, ha, he, ii, jej, jejku, o, och, oj, ojej, ojejku, ojoj, ou, u, uch, uf, cholera, Chryste Panie, holender, jasny gwint, Jezu, Jezus Maria, kurczę, kurde–mol, kur zapiał, psiakrew, rany boskie). To support this view, the scholar argues that they may substitute for each other (there are no specific syntactic or semantic contexts for their usage that would allow for any differentiation between them) and points out that it is impossible to determine any semantic representations for these expressions using the tools (in this case – the negation test) commonly applied for this purpose in structuralist semantics (Grochowski 1992: 159–160). The author also reviewed selected definitions of studied interjections recorded in “Słownik języka polskiego” edited by Witold Doroszewski and points out that they are based on open-ended enumerations of the properties of contexts in which the expressions being defined may occur, which does not allow for any delimitation of their hypothetical meanings (Grochowski 1992: 160–161). Grochowski classified such expressions as fe, fu, fuj, pfe, pfu, pfuj, hej, te, tss, tst, tssyt as meaning-bearing units in the class of inter-
jections proper, proposing preliminary semantic explanations for them (Grochowski 1992: 161–162). The proposed explications of expressions that Grochowski finds asemantic, such as o, oj, ojej, ach, och, may be found in Wierzbicka’s analyses (Wierzbicka 1991).
