Kai kurių lietuvių kalbos episteminio modalumo žymiklių ir modalinių frazių pagrindinis tonas ir intensyvumas

Acta Linguistica Lithuanica 2025, t. 93, p. 1–17 Contents link / Turinį nuoroda ISSN 2669-218X (online / elektroninis) FUNDAMENTAL FREQUENCY AND INTENSITY OF CERTAIN LITHUANIAN EPISTEMIC MODALITY MARKERS AND MODAL PHRASES Fundamental Frequency and Intensity in Lithuanian Epistemic Modality Markers and Modal Phrases EMILIJA CARBONI Vytautas Magnus University E-mail: [email protected] ORCID ID: https://orcid.org/0009-0007-5232-4479 Research area: phonetics. https://doi.org/10.35321/all93-03 ABSTRACT The study described in this article aims to identify the main characteristics of fundamental frequency and intensity in modal phrases: what fundamental frequency and intensity are characteristic of modal phrases, whether modality markers stand out in terms of fundamental frequency and intensity, and whether these features can signal different degrees of certainty in modal phrases. The research material consists of 17 sentences with lexical modality markers, one sentence in which modality is expressed phonetically, and one sentence neutral with respect to modality. These sentences were read three times each. The selected acoustic parameters (the mean, minimum, and maximum of fundamental frequency and intensity) were determined using the sound analysis program “Praat.” A total of 57 recordings were analyzed. The study applies experimental, instrumental sound-analysis, and descriptive methods. The results show that phrases simulating uncertainty have slightly higher fundamental-frequency parameters and lower intensity than neutral phrases of the same structure. Modal phrases with modality markers tend to have higher fundamental frequency and greater intensity than the neutral phrase, but the results are inconclusive. Modal words are characterized by higher fundamental frequency and greater intensity than the remainder of the phrase. Thus, they are emphasized, thereby accentuating the modality of the phrase. Phrases marking greater or lesser certainty are pronounced with greater intensity and lower fundamental frequency than phrases expressing doubt. KEYWORDS: intensity, modality, modality markers, fundamental frequency. Copyright © 2025 Emilija Carboni. Published by the Institute of the Lithuanian Language. This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. // Published by the Institute of the Lithuanian Language. This article is open access and distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the author and source are credited. Received: / Received: 2025–06–20. Accepted: / Accepted: 2025–12–05. 1 ABSTRACT The study aims to identify the key characteristics of fundamental frequency (pitch) and intensity in modal phrases: what pitch and intensity are typical of modal phrases, whether modal words are marked by pitch and intensity, and whether these features can signal varying degrees of certainty in modal phrases. The study material consists of 17 sentences containing lexical modal markers, along with one sentence expressing modality phonetically and one neutral sentence. Each sentence was read three times. The selected acoustic parameters for analysis (average, minimum, and maximum pitch and intensity) were determined using the sound analysis program “Praat”. A total of 57 recordings were analysed. The study employs experimental, instrumental acoustic analysis, and descriptive methods. The results indicate that phrases simulating uncertainty exhibit slightly higher pitch parameters and lower intensity compared to neutral phrases of the same structure. Modal phrases containing modal markers tend to have higher pitch and greater intensity than the neutral phrase, although the results may vary. Modal words are marked with a higher pitch and greater intensity than the rest of the phrase, thereby emphasising the modality of the phrase. Phrases expressing higher or lower certainty are pronounced with greater intensity and lower pitch compared to phrases expressing doubt. KEYWORDS: fundamental frequency, intensity, modality, modality markers. INTRODUCTION The category of modality continues to provoke considerable discussion in linguistic studies, which is why it is rather difficult to provide a precise definition of modality. This concept entered linguistics from the science of logic, now called modal logic (Latin modus – manner, harmony, mood), in which the so-called “modalities <...> are intended to define the truth of an utterance semantically” (Kaufmann et al. 2006: 71) (here and elsewhere translated by the article’s author); thus, logicians and philosophers define modality on the basis of the concepts of possibility / impossibility (English possibility / impossibility) and necessity (English necessity). In linguistics, modality is considered a semantic category of language. In some accounts, drawing on the ideas of logicians, modality is considered a “linguistic phenomenon that allows one to speak about situations that are not necessarily real by grammatical means” (Portner 2009: 1); in other words, it expresses the relation of an utterance to reality (how true or possible the utterance is). Elsewhere, scholars rely on the view of philosophers, according to which modality can also indicate the speaker’s attitude toward the content of a proposition (cf. Lyons 1977; Perkins 1983; Kaufmann et al. 2006 and others), i.e. modality is viewed as a linguistic means that makes it possible to express “the speaker’s evaluation” (Palmer 2003: 3), or that indicates an “attitude toward the situation being discussed” (Nuyts 2006: 1). According to Lithuanian linguists, modality, as a semantic category, is expressed in every sentence and indicates “the relation of the content to objective reality” (Sriubaitė 2011: 252). Although the views of some linguists differ, attempts have been made to systematize this category. In academic literature, linguistic modality is described by outlining its types and means of expression, and consequently various classifications exist. Theoretical approaches that examine modality from the speaker’s perspective distinguish two types: participant-internal (English participant-internal modality) and participant-external (English participant-external modality) modality, expressing a participant’s ability or need and the circumstances that make a situation possible or necessary (van der Auwera, Plungian 1998: 80). It is difficult to define all possible situations, and each of them may be expressed in various ways. Nevertheless, attempts have been made to do so; for example, Michaelis R. Perkins (1983), when describing modality, 2 relies on concepts distinguished in modal logic: alethic, deontic, temporal, epistemic, volitional, causal, evaluative, probabilistic modality. Sometimes researchers choose to look for specific modality markers, hoping that, for example, in English “the modality system is best illustrated by the functions of modal verbs”¹ (Palmer 2003: 2). According to Paul Portner (Portner 2009: 2), the most appropriate way to study modality is to determine the means by which it is expressed; thus, in the semantics of English, “auxiliary verbs <...>, adverbs <...>, adjectives <...>, which are most closely associated with the description of hypothetical situations, are analyzed first”². A synthesis of linguists’ views shows that every definition of modality leads to a twofold distinction: one modality indicates the speaker’s relation to the content of the utterance, while the other indicates the relation of the utterance’s content to reality. Thus, on this basis, academic literature distinguishes epistemic and non-epistemic modality. The first term is generally not disputed; it derives from the Greek word episteme, meaning knowledge or cognition, and epistemic modality consequently depends on the speaker’s knowledge (cf. Kaufmann et al. 2006; Portner 2009) or lack thereof (cf. Perkins 1983), for example, I am not sure whether these medicines cause side effects. The concepts already mentioned—certainty, uncertainty, doubt—are generally invoked for such an interpretation. This is the most widely used type of modality, which “indicates the degree of the speaker’s certainty about what is being said” (de Haan 2006: 29), and it can be expressed in various ways. In Lithuanian linguistics, the concept of epistemic modality is used similarly; it is stated that it indicates “the extent to which a particular speaker’s statement corresponds to reality and how certain the speaker is about what they are talking about” (Jasionytė-Mikučionienė 2015: 1). It is also recognized that, in actual language use, defining the speaker’s evaluation of reality is difficult, and it is noted that this “is a gradational evaluation, since its range encompasses both epistemic certainty and epistemic doubt (uncertainty). Naturally, there are many intermediate variants between these two extremes” (Usonienė 2016: 64). Non-epistemic modality is much more difficult to describe. In some accounts (de Haan 2006; Česnulienė 2012), it is simply called deontic modality (English deontic modality), but in most cases it is divided into deontic and dynamic (English dynamic modality) (Palmer 2003; Kaufmann et al. 2006; Nuyts 2006; Šolienė 2013 and others). Traditionally, deontic modality (Greek deon – obligation) expresses “the necessity or possibility for a morally responsible agent to perform certain actions” (Lyons 1977: 823). This type of modality therefore differs from epistemic modality in that it depends on external factors and does not characterize the content of an utterance; it “is based on a social or ethical norm” (Jasionytė-Mikučionienė 2016: 3), indicating what is “acceptable or unacceptable” (Portner 2009: 2), for example, You must study. Some studies distinguish dynamic modality, which, unlike deontic modality, indicates not an external requirement but the subject’s internal possibility or ability (cf. Palmer 2003; Nuyts ¹ „Modal systems are best illustrated by the functions of the modal verbs <...>“ (Palmer 2003: 2). ² „In semantics, this strategy has proceeded by first studying certain auxiliary verbs like must, certain adverbs like maybe, and certain adjectives like possible, since the meanings of these obviously have to do with situations that are not real“ (Portner 2009: 2). 3 2006), even if that subject is merely grammatical, for example, These medicines may cause side effects. Or dynamic possibility may also be determined by external circumstances, for example, He may escape (i.e. there are no obstacles to this) (cf. Palmer 2001: 77; van der Auwera, Plungian 1998: 80). Likewise, in Lithuanian linguistics, dynamic modality is understood as a possibility or necessity arising for any reason, when “a particular situation or action is described as possible or necessary because of the external circumstances conditioning it, the abilities or desire of the subject-participant” (Šolienė 2013: 28). Other modality subtypes and classification methods also exist, for example, the alethic modality discussed by Jan Nuyts (2006), the discourse, sentence, constituent modality mentioned by P. Portner (Portner 2009), or the participant-internal / participant-external modality of Johan van der Auwera and Vladimir Plungian (van der Auwera, Plungian 1998), but these are neither named nor analyzed in greater detail in the works of other authors. The main types of modality discussed above are usually distinguished by lexical means of expression. The most common are modal adverbs and particles (for example: perhaps, maybe, probably, certainly), participles (for example: possible, likely, plausible), mental-state verbs (English mental state predicate) (for example: to think, to believe), auxiliary verbs (for example: can, have to, must), and nouns (for example: possibility, probability) (cf. Nuyts 2001; Šolienė 2013; Usonienė 2016). Moods and tenses are considered grammatical means of expression, for example, These medicines could cause side effects. In English, modal verbs are most commonly used to mark epistemic modality (Šolienė 2007); in the works of some scholars (Perkins 1983 and others), they are also called primary modality markers, for example, can (English can / may), must (English must), or will, indicating a future action. Lithuanian is much more flexible in this respect, and modal verbs may be replaced by particles or adverbs, for example, perhaps / most likely / maybe these medicines cause side effects. The analysis of modality in English already has a tradition, with more theoretical works and empirical studies. However, not all languages have obvious manifestations or equivalents of modality. For example, Italian “has neither specific markers for expressing epistemic modality nor clear grammatical forms” (Pietrandrea 2005: 61); the main modality markers in Italian are considered to be the verbs must (Italian dovere) and can (Italian potere), more precisely, their future-tense and subjunctive forms (ibid.). Although the expression of modality may differ across languages, it is believed that “in languages that do not have a separate classification of modality markers, other linguistic means are used to express modality: certain adverbs, intonation, word order in the sentence, and specific use of moods” (Favaro 2023: 22). Thus, modality can be regarded as a universal language category, although its expression differs across languages. Most studies on modality in Lithuanian consist of comparative studies of Lithuanian and English (Usonienė 2006a; 2006b; Šolienė 2007; 2013; 2015; Šinkūnienė, van Olmen 2012; 2015), usually based on examples from parallel corpora. Such studies are useful for determining what possibilities for expressing modality exist in Lithuanian in general. For example, the problem has been raised that, in the Dictionary of the Lithuanian Language (LKŽe), probably 4 is equated with perhaps and most likely, which seems to suggest that the concept of probability is not measured in Lithuanian, although intuitively the difference is obvious” (Usonienė 2006a: 100). Although most attention has been devoted to the analysis of specific markers, it has been observed that the entire certainty/uncertainty continuum is particularly important for the expression of epistemic modality: “the expression may be varied, lexical or grammatical; what matters is that the degree of probability does not change” (ibid.). Such studies are presumably useful not only to lexicographers but also important for translators, because they help provide a better understanding of less extensively discussed language categories. For example, it has been observed that “in translations, markers of epistemic necessity are replaced by particles of epistemic certainty, disregarding the degree of certainty or possibility” (Šolienė 2007: 98). A review of the work of Lithuanian scholars suggests that epistemic modality has so far received more attention. Another important aspect of modality research in Lithuanian is the variety of discourses. Quantitative studies show that the most frequent particles maybe and perhaps “appear quite frequently in literary texts, but their prototypical use is characteristic of spoken language. <...> [whereas] authors of scholarly texts use other linguistic means to express their attitude as authors and tend to assume responsibility for the truth of their claims” (Šolienė 2015: 156). Nevertheless, more detailed qualitative studies do not confirm this observation: in scholarly language, authors do not avoid modality and do not always write confidently (Šinkūnienė 2015). These conclusions draw attention to the importance of practical research and show that actual usage does not always confirm the theoretical classification of modality; for example, “the distinction between the deontic and non-deontic subtype of necessity is not always clear and depends on context” (Jasionytė-Mikučionienė 2016: 4). The studies of modality in Lithuanian that have been conducted are significant; nevertheless, it should be noted that Lithuanian linguistics has so far examined only written language. It is therefore important to expand the research material, encompass as many means of expression as possible, and, of course, take into account the context in which modality markers are used. As mentioned above, considerable work has been devoted to the expression of modality, but possible phonetic features of modality have not been studied either in Lithuanian or in other languages. Some works mention the importance of intonation (Sriubaitė 2011; Česnulienė 2012; Favaro 2023), and one study of spoken Italian states that “modality in spoken language may also be indicated by other features, such as prosodic elements, facial expression, gestures, etc.” (Tucci 2011: 87). Nevertheless, none of these articles explains which specific features of intonation or prosody mark modality or how they do so. Another aspect that complicates the study of phonetic features is that modality may also be evaluated from a pragmatic perspective, when modality markers acquire epistemic, deontic, and other meanings only in a particular context and depend on the interlocutor’s choice, linguistic experience, and variety of expression. However, collecting representative, high-quality examples of spontaneous speech containing a sufficient number of different modal words is difficult; moreover, the speaker’s intention would still remain unknown, or at best only inferable. There are works that still attempt to draw clear boundaries between language categories. For example, a study devoted to illocution (English illocution), modality, and attitude (English attitude) states that some authors, “for example, Ungeheuer (1972), [speech acts and modality] regarded as the same category, <...> whereas Lüdtke (1980) <...> [regarded them as] simply different types of the author’s attitude toward the content of the utterance 5 (Mello, Raso 2011: 3). Thus, it is still unclear when certain lexical units in written language can be considered modality markers by which the author expresses uncertainty, possibility, or probability, and when they express a particular illocution, for example, when one wishes to inform, remind, authorize, and so on. It is sometimes stated that “both modality and illocution indicate the author’s ‘attitude’” (Tucci 2011: 85)3. The same applies to manner of utterance. According to some authors, “some manners of utterance are related to the content of the proposition (for example, irony, uncertainty, obviousness, surprise, etc.), whereas others are related to the social relations between interlocutors during the communicative act (for example, politeness, arrogance, authority, etc.)” (Moraes et al. 2010, cited in Mello, Raso 2011: 4). Consequently, in addition to lexical and grammatical expression, other distinctive—or, more precisely, psychoacoustic—features of similar categories should be identified. It is presumed that research into phonetic expression would help define not only these concepts more precisely, since “because both hedges and epistemic modality express the speaker’s attitude toward the truth of the proposition uttered, these two phenomena are closely related” (Šinkūnienė 2008: 98). The novelty of the research currently prevents answers to all the questions raised here and similar questions; therefore, the empirical data presented in this article cannot yet be compared with studies of foreign languages. Nor are important conclusions drawn about the place of modality among other semantic categories; the analysis is limited to markers of one type of modality in sentences of similar structure, in the hope that the results obtained will be useful in further research. The aim of the study is to identify the main characteristics of fundamental frequency and intensity in modal phrases: what fundamental frequency and intensity are characteristic of modal phrases, whether modality markers stand out in terms of fundamental frequency and intensity, and whether these features can signal different degrees of certainty in modal phrases. The object of the study is the duration (tempo) of speech-sound units, fundamental frequency, and intensity in phrases of read speech expressing modality (hereinafter, modal phrases). The research material consists of a baseline phrase that is semantically free of emotional coloring—These medicines increase blood pressure—that is, a sentence read with neutral intonation, with modal words added to it. In the study, these are assigned to two groups: those marking uncertainty or doubt (hardly, I am not sure, I wonder, maybe, probably, perhaps, it seems, it would seem, apparently) or varying degrees of certainty (most likely, likely, probably, undoubtedly, of course, most importantly, unfortunately, regrettably; the latter have a shade of regret). The modality-expressing units were selected according to the list provided in the Functional Grammar of the Lithuanian Language (see Valeckienė 1998: 76–77). After a preliminary study with several speakers, the question was raised whether the results could have been influenced by contrastive stress, because the modal words were pronounced with higher pitch and greater intensity; why some units were emphasized more; and whether the speakers might have failed to recognize some modality markers, since some of them (for example, may, could) occurred in the middle of the sentence and were not related to the speaker’s evaluation of the content of the utterance. Therefore, this study uses not the predicate with the prefix—causes—but the predicate increases; it is treated as the central 3 „<…> both modality and illocution communicate a speaker’s ‘attitude’“ (Tucci 2011: 85). 6 word of the phrase (it always receives contrastive stress), thus allowing observation of whether intonation changes in any way because of the modality marker at the beginning. Modality markers in the middle of the sentence were also excluded. 1. METHODOLOGY FOR STUDYING THE FUNDAMENTAL FREQUENCY AND INTENSITY OF MODAL PHRASES When constructing the sentences, it was decided to rely only on examples of epistemic modality, because these most clearly reflect subjectivity in language and are the most frequently used; their phonetic features could therefore be more pronounced. Each sentence was read three times. In addition, the baseline sentence was read three times not only neutrally but also in an attempt to imitate modality (uncertainty, doubt). The recordings were produced by the article’s author (24 years old), in order to avoid considerable data variation and ensure the reliability of the results: while reading the sentences, the same distance between the microphone and the mouth was maintained, and the phrases were pronounced while attempting to convey as accurately as possible the meanings marked by the modal units; the same central phrase word was maintained throughout. This decision was also motivated by experience from an earlier study, in which speaker variation did not help reveal possible regularities. Therefore, it was expected that if, after recording with knowledge of the research object, no pronounced acoustic differences were found, it could be predicted that modality is expressed semantically and grammatically but not phonetically. A total of 57 recordings were analyzed: 51 recordings of phrases with lexical modality markers, 3 recordings of phrases in which modality was expressed only through intonation, and 3 recordings of baseline phrases. The sound-analysis program “Praat” (Boersma, Weenink 2022, version 6.2.23) was used to segment and annotate the recordings and determine the selected phonetic features. First, the recording was listened to and segmented (the boundaries of phrases and words were marked). Figure 1 presents the program’s main analysis window, showing the sound wave at the top, the spectrogram in the middle band, and the annotated phrase at the bottom. Figure 1. Main window of the “Praat” program 7 Next, the segment under study was selected in each recording: the entire phrase, the modal word, and the remainder of the phrase. For these phrase portions, the mean and maximum fundamental frequency (F0) and intensity, as well as the standard deviation and minimum fundamental frequency, were measured. The statistical significance of the data was assessed using the t criterion (the significance level is 0.5). The study uses experimental, instrumental sound-analysis, and descriptive methods. 2. MODALINIŲ FRAZIŲ IR MODALUMO ŽYMIKLIŲ PAGRINDINIS TONAS Fundamental frequency (English fundamental frequency), denoted by F₀, indicates the frequency of vibrations produced by the vocal folds. In this study, F₀ is measured in hertz (Hz). The program provided the mean fundamental frequency of the segments under study (F₀mean), its highest (F₀max) and lowest (F₀min) values. The standard deviation of F₀, or standard deviation (hereinafter F₀SD), indicates how much the fundamental frequency varies; that is, the wider the fundamental-frequency range and the more the pitch varies during speech, the higher the resulting F₀SD. The results for mean fundamental frequency (F₀mean) show that in 7 of the 18 modal phrases, F₀mean is higher than in the neutral phrase. No consistent differences in F₀mean were found among modal phrases marking different degrees of uncertainty; the F₀mean of phrases marking doubt and certainty was nearly identical: 215 Hz and 212 Hz, respectively. It is noteworthy that the phrase without a modality marker but uttered while imitating uncertainty has a relatively high F₀mean and is among the five phrases with the highest F₀mean (see Table 1). Table 1. F₀mean of phrases and modality markers, F₀min and F₀max of phrases (Hz) 8 Most importantly, these medicines increase blood pressure. 208 (±24) 223 204 168 302 Unfortunately, these medicines increase blood pressure. 209 (±25) 215 208 166 281 Regrettably, these medicines increase blood pressure. 212 (±24) 220 210 173 285 In all 17 phrases with modality markers, the stationary fragment (These medicines increase blood pressure) was uttered at a lower pitch than the modal unit, but the difference varies considerably. For example, in phrases with maybe, perhaps, it seems, most likely, the difference is very small, and it can be stated that the modality marker was not distinguished by fundamental frequency either. The modality marker’s F₀mean was on average 1.1 times higher than that of the remainder of the phrase; these differences were statistically significant (tₚ=5,4>t₀,₀₅=2,1). Consequently, modality markers are emphasized and probably receive phrasal stress (English phrasal stress)—they become the most important word in the phrase and are highlighted by F₀, or become an independent intermediate phrase (English intermediate phrase), i.e. a constituent of the intonational phrase. The F₀mean of the marker maybe is most similar to that of the stationary fragment of the same phrase (These medicines increase blood pressure). This is apparently influenced by the small number of sounds and the fact that it consists of only one syllable; therefore maybe becomes a clitic and attaches to the following word, and the entire phrase is pronounced with a fairly high fundamental frequency. When comparing both the modal unit expressing doubt and the one expressing certainty, as well as the F₀vid of the stationary fragments following them respectively, no more pronounced differences were observed. Although the F₀vid of the modal units expressing doubt and of the remaining part of the phrase (232 Hz and 210 Hz, respectively) are slightly higher than those of the phrases expressing certainty (228 Hz for the modal units and 207 Hz for the remaining part of the phrase), the differences between the corresponding parts of the phrases expressing doubt and certainty are not statistically significant (for the modal units, t criterion tₚ=0,58<t₀,₀₅=2,1; for the remaining part of the phrase – tₚ=0,84<t₀,₀₅=2,1). Studies of emotions in the Lithuanian language (Dereškevičiūtė, Kazlauskienė 2016) established that, when expressing stronger emotions, the values of the standard deviation of the fundamental frequency (F₀SN) increase. It was predicted that in this study differences in F₀SN could be determined by the modal word: if modality is indeed related to (un)certainty, neutral phrases would have a lower F₀SN. Only three phrases have an F₀SN identical to that of the reference phrase; in the other phrases, F₀SN is on average 1.3 times higher. The F₀SN of phrases with modal markers vargu, nesu tikra, gal, tikėtina, tikriausiai, rodos, matyt, žinoma differs most from that of the neutral phrase. As can be seen, this list includes modal units expressing both considerable and slight doubt, which shows that F₀SN may vary even when speaking confidently; for example, when confidently reading a sentence with the marker žinoma, the voice pitch varies almost to the same extent as when reading a sentence with vargu. Moreover, the F₀vid of these two units is particularly high relative to the rest of the phrase. In the phrases studied, the maximum fundamental frequency (F₀max) ranges from 275 Hz to 333 Hz. The highest F₀max values were recorded in phrases with the modal markers tikėtina, matyt, žinoma, vargu (319–333 Hz). It can be seen that a higher F₀max is not related to differences in the meaning of the modal unit 9 among them are the high-probability žinoma and the completely opposite vargu. Thus, no difference in F₀max was found between phrases expressing doubt and certainty; the means of both groups were identical: 303 Hz and 302 Hz, respectively. In this study, no difference in F₀min was established between phrases expressing doubt and certainty; the means of both groups were identical: 168 Hz and 166 Hz, respectively. It should be noted that the F₀max of the neutral phrase and the phrase imitating uncertainty are among the lowest; three further phrases with modal markers greičiausiai, deja, gaila have a similar F₀max. In the latter phrases, the modal words were uttered with the lowest fundamental frequency; they are emphasized the least, which results in a low F₀max. On the other hand, the F₀min of all phrases with modal markers is lower than the F₀min of the neutral phrase and the phrase imitating uncertainty. One of the most pronounced differences was observed when analyzing neutral phrases and phrases expressing uncertainty without a modal marker: in the neutral phrase, F₀ is highest at the beginning; the fundamental frequency rises on the first syllable of the word, then drops sharply and remains low (see Figure 2), whereas in the phrase in which modality is attempted to be expressed through the voice alone, F₀ rises at the predicate, then decreases slightly but remains high throughout the end of the phrase (see Figure 3). Figure 2. F₀ curve of the neutral phrase (150–300 Hz) 10 Figure 3. F₀ curve of the phrase uttered with uncertainty (150–300 Hz) A different F₀ tendency can be seen in the curves at the beginning of phrases with modal words. In such phrases, only the modal word is uttered with a higher fundamental frequency; the F₀ characteristic of the remaining part of the phrase is the speaker’s usual F₀ (see Figure 4). Thus, the modal marker receives logical stress, thereby reinforcing the modality of the phrase, which is primarily expressed lexically and grammatically. Figure 4. F₀ curve of a phrase with the modal marker vargu (150–350 Hz) As can be seen in the figure, the fundamental-frequency curve reaches its highest point at the modal word and immediately falls afterward—the modal word is clearly emphasized. This separation of the modal word from the entire phrase, and the subsequent slight emphasis of the central phrase word didina, shows that modal phrases may be uttered as two internal phrases: one centered on the modal word and the other on the predicate didina. In all 17 sentences with modal markers, the stationary fragment was uttered at a lower pitch than the modal unit. 11 Among the fundamental-frequency parameters analyzed, the most pronounced regularities can be associated with the F0vid of the modal markers, which is statistically significantly higher than the F0vid of the remaining part of the phrase. In this study, phrases with modal markers are characterized by higher F0max and lower F0min than the neutral phrase or the phrase imitating uncertainty, but because of the small sample of the latter phrases, the difference cannot be evaluated statistically. 3. MODALINIŲ FRAZIŲ IR MODALUMO ŽYMIKLIŲ INTENSYVUMAS Intensity (English intensity), or strength, is also part of the set of acoustic features; it indicates the strength with which the vocal folds vibrate during speech. In this study, intensity is measured in decibels (dB). Of the 18 modal phrases, 13 were read with greater intensity than the reference phrase. The lowest mean intensity value is characteristic of the uncertainly read phrase without a modality marker. The phrases with the markers gal, greičiausiai, nesu tikra, tikėtina, turbūt differ least from the reference phrase; their mean intensity ranges from 47,06 to 48,34 dB, whereas the mean intensity of the other phrases is considerably higher, between 60,62 dB and 61,91 dB (see Table 2). Table 2. Mean intensity of phrases and modal words (INTvid) and maximum intensity of phrases (INTmax) (dB) Sentence | INTvid of all phrases | Modal word INTvid | INTvid of phrase part | Phrase INTmax These medicines increase blood pressure. (neutral phrase) | 49,16 | - | - | 55,26 These medicines increase blood pressure. (simulated modality) | 45,79 | - | - | 51,81 Hardly, these medicines increase blood pressure. | 61,09 | 63,32 | 60,26 | 68,14 I am not sure whether these medicines increase blood pressure | 47,18 | 48,98 | 46,36 | 55,55 I wonder whether these medicines increase blood pressure. | 61,39 | 63,86 | 61,43 | 68,65 Perhaps these medicines increase blood pressure. | 47,06 | 51,18 | 49,04 | 52,68 In all likelihood, these medicines increase blood pressure. | 61,91 | 64,22 | 61,03 | 68,28 These medicines probably increase blood pressure. | 48,34 | 48,80 | 48,09 | 54,63 It seems these medicines increase blood pressure. | 61,06 | 62,06 | 60,82 | 67,32 Apparently, these medicines increase blood pressure. | 60,62 | 62,25 | 59,89 | 67,16 These medicines evidently increase blood pressure. | 61,54 | 62,46 | 61,26 | 67,79 These medicines are most likely to increase blood pressure. | 46,89 | 47,10 | 46,71 | 52,96 12 Likely, these medicines increase blood pressure. 47,94 47,19 48,25 53,80 Most likely these medicines increase blood pressure. 61,45 62,61 60,99 68,80 Of course, these medicines increase blood pressure. 62,48 65,59 60,71 68,95 Certainly, these medicines increase blood pressure. 60,76 62,84 59,81 66,88 Most importantly, these medicines increase blood pressure. 62,26 63,34 61,75 68,66 Unfortunately, these medicines increase blood pressure. 61,67 63,77 60,68 67,67 It is a pity, these medicines increase blood pressure. 62,87 65,59 61,95 69,13 The mean intensity depends little on the type of modality: both phrases expressing doubt and those expressing certainty may be uttered more forcefully. Phrases expressing greater or lesser certainty are characterized by greater intensity (58,29 dB on average) than phrases expressing doubt (56,69 dB). However, because of the large dispersion of values in both groups, the difference is not statistically significant (tₚ=0,48<t₀,₀₅=2,1). All modal markers have more or less greater intensity than the remaining part of the phrase. The modal markers were uttered 1.6 times4 more forcefully than the entire remaining phrase, although because of the large dispersion of values the difference is not statistically significant (tₚ=0,93<t₀,₀₅=2); nevertheless, it should be acknowledged that other statistical criteria should be sought in the future to establish the statistical significance of intensity differences. The intensity of modal words expressing varying degrees of certainty is, in general, 1.3 times greater than that of modal words expressing doubt. The highest mean intensity was established for the markers in all likelihood, of course, it is a pity. Their mean fundamental frequency was also among the lowest, although higher relative to the stationary fragment; therefore, it can be assumed that the lower fundamental frequency is compensated for by greater intensity. When the modality markers were ranked by mean intensity, the first five with the highest intensity values, in addition to those already mentioned, of course, it is a pity, in all likelihood, are kažin and unfortunately. The lowest F₀vid was established for unfortunately, probably, most likely, in all likelihood, it is a pity. As can be seen, three of them were read more intensely. The other two markers, most likely and probably, were characterized by both low F₀ and low intensity. The modal word likely, which was most distinct in terms of F₀vid, was read with the lowest intensity, whereas none of the words with the highest F₀vid—likely, apparently, hardly, most likely—had a high mean intensity. The intensity and its maximum of modal phrases expressing certainty are 1.3 times higher than those of phrases expressing doubt. The lowest maximum-intensity value ⁴ The intensity ratio was calculated differently from that of fundamental frequency, where one value is divided by the other. If the overall mean intensity of the modal marker is 59,13 dB and that of the remaining part of the phrase is 57,002 dB, the intensity ratio is 10⁽⁵⁹,¹³⁻⁵⁷,⁰⁰²⁾/¹⁰=1,6. 13 is characteristic of the uncertainly uttered phrase without a modal marker. When checking where the curve reaches its highest point, the same result can be seen in all recordings—the greatest intensity is reached when uttering the predicate, where the fundamental-frequency curve also rose. The question may arise why the intensity values of these phrases are lower and the fundamental frequency higher than those of the reference phrase if the same word is emphasized in both phrases. The essential difference between these phrases lies in their final acoustic features. To read the sentence expressing uncertainty, the end of the phrase is prolonged, the tempo slows, intensity decreases, whereas the fundamental frequency, conversely, rises. According to all the intensity parameters analyzed, modal phrases expressing greater or lesser certainty and their modal markers are uttered more intensely than phrases expressing doubt and their modal markers. The phrase imitating uncertainty has the lowest intensity; its maximum intensity is also the lowest. CONCLUSIONS After investigating fundamental frequency and intensity in a phrase neutral from the perspective of modality, a phrase imitating uncertainty, and modal phrases with modality markers, the following conclusions can be drawn. The fundamental-frequency parameters examined in the neutral phrase and the phrase imitating uncertainty are similar. The mean and minimum fundamental frequencies of the neutral phrase and the phrase imitating uncertainty are similar; no obvious tendencies were established. Modal words are uttered with a higher fundamental frequency than the remaining part of the phrase, while the entire modal phrase differs from the neutral phrase by a higher maximum fundamental frequency. Thus, in a modal phrase, the modality marker is emphasized and receives logical stress or becomes an independent internal phrase. The phonetic features of the modal word and its status in the structure of the prosodic phrase should be studied in greater detail. The highlighting of the modal word also determines a greater standard deviation of the mean fundamental frequency of modal phrases. Although all the fundamental-frequency parameters studied indicate that phrases expressing doubt tend to have a higher fundamental frequency, the differences are too small to permit reliable conclusions. The mean and maximum intensity of the phrase imitating uncertainty are the lowest in the material studied. Thus, low intensity could be one of the features of expressing modality through the voice, but more research is needed to confirm this. Modal phrases with modal markers tend to have greater intensity than the neutral phrase, but the results are ambiguous. However, the maximum intensities of these phrases are not always greater than that of the neutral phrase. The modal markers in the research material were uttered with greater intensity than the remaining part of the phrase. All the intensity parameters studied show that phrases expressing greater or lesser certainty were uttered more intensely than phrases expressing doubt. In summary, it can be stated that: 1) the fundamental-frequency parameters of the phrase imitating uncertainty are slightly higher, and its intensity lower, than those of a neutral phrase of the same structure; 2) modal phrases with modal markers tend to have a higher fundamental frequency and greater intensity than the neutral phrase, but the results are ambiguous; 14 3) modal words are characterized by a higher fundamental frequency and greater intensity than the remaining part of the phrase; they are emphasized, thereby highlighting the modality of the phrase; 4) phrases expressing greater or lesser certainty are uttered with greater intensity and a lower fundamental frequency than phrases expressing doubt. The emerging tendencies suggest that modality in read speech may be expressed not only by lexical or grammatical means but also by acoustic means of expression. These results indicate directions for further research: increasing the sample, inviting more speakers, and preparing more sentences (taking into account the position of the modality marker in the sentence and nuances of its meaning); analyzing more acoustic features (frequency and amplitude fluctuations, duration, etc.); and revising the division of modality markers into only two groups. It must be noted that the expression of modality may depend on the speaker, so a larger data sample is required for a more comprehensive study. However, relying on recordings of spontaneous speech is difficult because it is not always clear why speakers choose one or another modality marker. Controlled speech should therefore be studied and participants in conversations surveyed. In addition, it remains unclear whether modality and uncertainty are in fact equivalent concepts, or whether the phonetic features of interactional language, illocution, and manner of utterance would differ. LITERATŪRA Boersma Paul, Weenink David 2022: Praat: Doing Phonetics by Computer, version 6.2.23. Prieiga internete: https://praat.org. Česnulienė Vida 2012: Modaliniai konektoriai mokslinio ir publicistinio stiliaus tekstuose [Modal Connectors in the Texts of Scientific and Publicistic Style]. – Žmogus ir žodis 14(1), 30–34. de Haan Ferdinand 2006: Typological Approaches to Modality. – The Expression of Modality, ed. W. Frawley, Berlin, New York: Mouton de Gruyter, 27–69. DOI: https://doi.org/10.1515/9783110197570.27. Dereškevičiūtė Sigita, Kazlauskienė Asta 2016: Emocijos: jų atpažinimas ir kai kurie akustiniai požymiai [Emotions: Recognition and Some Acoustic Features]. – Baltistica, 51, 129–148. DOI: 15388/baltistica.51.1.2242. Favaro Marco 2023: Modal Particles in Italian: Adverbs of Illocutionary Modification and Sociolinguistic Variation, Berlin: Language Science Press. DOI: https://doi.org/10.5281/zenodo.10259474. Jasionytė-Mikučionienė Erika 2015: Modalumo tyrimas senuosiuose XVI–XVII a. lietuvių kalbos raštuose [A Study of Modality in Old 16th–17th Century Lithuanian Writings]. – Gimtoji Kalba 9, 12–14. Jasionytė-Mikučionienė Erika 2016: Semantinės ir struktūrinės modalumo raiškos priemonių ypatybės senuosiuose (XVI a.) lietuvių kalbos raštuose [Semantic and Structural 15 Properties of Modal Markers in Old Lithuanian Writings (the 16th Century)]. – Lietuvių kalba 10, [1–22]. DOI: http://dx.doi.org/10.15388/Lietkalb.2016.10.9926. Kaufmann Stefan, Condoravdi Cleo, Harizanov Valentina 2006: Formal Approaches to Modality. – The Expression of Modality, ed. W. Frawley, Berlin, New York: Mouton de Gruyter, 71–106. Mello Heliana, Raso Tommaso 2011: Illocution, Modality, Attitude: Different Names for Different Categories. – Pragmatics and Prosody: Illocution, Modality, Attitude, Information Patterning and Speech Annotation, ed. H. Mello, A. Panunzi, T. Raso, Firenze: Firenze University Press, 1–18. Nuyts Jan 2001: Epistemic Modality, Language, and Conceptualization: A Cognitive–Pragmatic Perspective, Philadelphia: John Benjamins Publishing Company. DOI: https://doi.org/10.1075/hcp.5. Nuyts Jan 2006: Modality: Overview and Linguistic Issues. – The Expression of Modality, ed. W. Frawley, Berlin, New York: Mouton de Gruyter, 1–26. Palmer Frank R. 2001: Mood and Modality, 2nd ed., Cambridge: Cambridge University Press. Palmer Frank R. 2003: Modality in English: Theoretical, Descriptive and Typological Issues. – Modality in Contemporary English, ed. by R. Facchinetti, F. Palmer, M. Krug, Berlin, New York: Mouton de Gruyter, 1–17. DOI: https://doi.org/10.1515/9783110895339.1. Perkins Michael R. 1983: Modal Expressions in English, London: Frances Pinter. DOI: https://doi.org/10.2307/414408. Pietrandrea Paola 2005: Epistemic Modality: Functional Properties and the Italian System, Amsterdam, Philadelphia: John Benjamins Publishing Company. Portner Paul 2009: Modality, Oxford: Oxford University Press. Lyons John 1977: Semantics, Cambridge: Cambridge university press. Sriubaitė Jurgita 2011: Text Modality and the Modus in Lithuanian and English art Review. – Kalba ir kontekstai 2, 249–257. Šinkūnienė Jolanta 2008: Autoriaus pozicijos švelninimas: tarpdal ykiniai ir tarp kalbiniai raiškos priemonių ypatumai [Hedging in Lithuanian and English Research Articles: A Cross – Disciplinary and Cross – Linguistic Study]. – Kalbotyra 58, 97–108. Šinkūnienė Jolanta 2015: Neepisteminis modalumas lietuvių ir anglų mokslo kalboje: kiekybiniai ir kokybiniai vartosenos ypatumai [Nonepistemic Modality in English and Lithuanian Academic Discourse: Quantitative and Qualitative Perspectives]. – Kalbotyra 67, 131–154. DOI: http://dx.doi.org/10.15388/Klbt.2015.8946. Šinkūnienė Jolanta, van Olmen Daniel 2012: Modal Verbs of Necessity in Academic English, Dutch, and Lithuanian: Epistemicity and/or Evidentiality? – Darbai ir dienos 58, 153–181. 16 Šolienė Audronė 2007: Epistemic Modality in English and Lithuanian Translations. – Kalba ir kontekstai 2, 90–102. Šolienė Audronė 2013: Episteminio modalumo ekvivalentiškumo parametrai anglų ir lietuvių kalbose: daktaro disertacija [Parameters of Epistemic Modality Equivalence in English and Lithuanian: Doctoral Dissertation], Vilnius: Vilniaus universitetas. Šolienė Audronė 2015: Multifunctionality of Modal Markers: Lithuanian Epistemic Adverbials gal and galbūt ‘perhaps/maybe’ vs. Their Translational Correspondences. – Kalbotyra 67, 155–176. DOI: http://dx.doi.org/10.15388/Klbt.2015.8947. Tucci Ida 2011: Illocution and Modality in Spoken Italian: Performing a Speech Act Through Words and Judging Their Semantic Content: a Corpus-Based Analysis. – Pragmatics and Prosody: Illocution, Modality, Attitude, Information Patterning and Speech Annotation, ed. H. Mello, A. Panunzi, T. Raso, Firenze: Firenze University Press, 83–100. Usonienė Aurelijia 2006a: Episteminio modalumo raiška: ekvivalentiškumo bruožai anglų ir lietuvių kalbose [The Expression of Epistemic Modality: Equivalence Features in English and Lithuanian]. – Darbai ir dienos 45, 97–108. Usonienė Aurelijia 2006b: Tikimybės veiksmažodžiai anglų ir lietuvių kalbose: atitikmenų paieška [Verbs of Probability in English and Lithuanian: In Search of Equivalents]. – Baltistica, 41, 321–332. DOI: http://dx.doi.org/10.15388/baltistica.41.2.1002. Usonienė Aurelijia 2016: Reikšmės pasaulis: tekstynais paremti semantiniai kalbų tyrimai [The World of Meaning: Text-Based Semantic Language Research], Vilnius: Vilniaus universitetas. Valeckienė Adelė 1998: Funkcinė lietuvių kalbos gramatika. Kalbos ugdymo programa [Functional Grammar of the Lithuanian Language. Language Education Program], Vilnius: Mokslo ir enciklopedijų leidybos institutas. van der Auwera Johan, Plungian Vladimir 1998: Modality’s Semantic Map. – Linguistic Typology 2(1), 79–124. 17