No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
The OCP and gradient data
Abstract
The abstract for this document is available on CSA Illumina.To view the Abstract, click the Abstract button above the document title.
... This paper addresses the question of gradient prohibition against similar consonants in proximity to each other (OCP effect) in English phonotactics. Previous researchers (Berkley 1994b) established that in English consonant pairs that share place of articulation are underrepresented in monosyllabic monomorphemic words. In addition, she found that the strength of this prohibition in English was dependent on the amount of distance between the two consonants, quantified as the number of segments intervening between the consonants. ...
... In addition, we go further in examining the distance effect by hypothesizing that vowel height may act as another factor conditioning the strength of the OCP. Berkley (1994b) showed that the amount of intervening material matters for the strength of the OCP. Vowels of different height are intrinsically different in terms of length and intensity. ...
... In other words, similarity is in inversely related to the strength of the restriction. It was also discovered that the elements do not have to be strictly adjacent (McCarthy 1988, Berkley 1994b, Frisch et al. 2004, Pater and Coetzee 2005, Coetzee and Pater 2008. For example, McCarthy (1986a) showed that in Arabic homorganic consonants are prohibited even when they are separated by a vowel as a result of a morphological derivation, although it is assumed that on the level of representation they belong to the same tier, and are, therefore, adjacent. ...
This paper focuses on the factors affecting gradient well-formedness of English CVC syllables. The study
examines a gradient OCP-place effect in English CVC words and syllables extracted from two electronic
dictionaries, CMU and CELEX. It was found that a gradient restriction on the co-occurrence of homorganic
consonants operates in all CVC syllables of English. Previously this was established only for monomorphemic
monosyllabic words (Berkley 1994). The distance effect, reported by Berkley (1994), was only partially confirmed
in this study. Syllables containing long vowels or diphthongs in many cases exhibited a weaker OCP effect than
those containing short vowels, but this difference was only marginally significant. The effect of vowel height was
also examined as a possible factor affecting the strength of the OCP. The results showed a non-significant trend in
the predicted direction. Another factor affecting the well-formedness of the CVC syllables is the prominence
alignment between syllable stress, vowel height, and consonant place. Stressed syllables combine more often with
more sonorous low vowels. Unstressed syllables prefer less sonorous high and reduced vowels and coronal onsets
and codas. It was established that the syllables violating these requirements were under-represented in the lexicon.
Following Coetzee and Pater (2008) we fit a linear regression model to the data and propose an OT approach
based on partially ordered grammar (Anttila 1998). Both the statistical model and the OT grammar feature
constraints capturing OCP related restrictions and prominence alignment related restrictions on the well-formedness
of English CVC syllables. The proposed OT account provides a reasonably close fit to the quantitative patterns
observed in English phonotactics.
... In other words, the acceptability of a form is gradient and the acceptability of a form can be measured by its lexical frequency. Berkley (1994b), Pierrehumbert (in press), and Plenat (1996) argue that accounting for differences in lexical frequency for otherwise equivalent forms is impossible in a non-quantitative phonological formalism. We propose that gradient acceptability be accounted for by gradient phonological constraints. ...
... Note that these phonotactic restrictions are part of a native speaker's implicit linguistic knowledge, and not merely artifacts of historical ancestry. OCP-Place effects are reflected in language games (McCarthy 1986), in morphological derivation (Berkley 1994b, Bybee and Slobin 1982, Jespersen 1939, Liberman 1994, and in the interpretation of complex words (Pierrehumbert 1994). Therefore, gradient constraints should be included in our model of phonological competence as part of the theory of synchronic grammar. ...
... However, we note that other cases are known in which OCP effects must apply to surface representations. Berkley (1994b) shows that English OCP effects apply across morpheme boundaries and are gradiently sensitive to distance. Recall that OCP-Tone effects are also sensitive to prosodic and morphological structure (Odden 1986, Pierrehumbert andBeckman 1988). ...
The statistical patterns of language are systematic linguistic data which must be accounted for in linguistic theory. The phonotactics of the Arabic verbal roots are presented as a case study. In Arabic, the acceptability of a verbal root is gradiently dependent on the similarity of homorganic consonant pairs within the root. We propose the stochastic constraint model of phonological patterning, in which the relative frequency of a phonological form provides a measure of its acceptability. The stochastic constraint model can be parameterized to account for gradient or categorical constraints. We also propose a novel similarity metric for phonological segments based on the representational framework of structured specification . Structured specification provides a basis for a similarity metric for phonemes that is sensitive to featural redundancy and contrastiveness. This metric is superior to previous proposals using underspecification. The account of Arabic consonant cooccurrence using similarity and the stochastic constraint provides a more accurate account of the data than the non-quantitative autosegmental account, demonstrating that a quantitative description is necessary to capture the true pattern of the data.
... That is, the net result of adjudicating constraint violations is the same for all input forms that share the relevant particulars. In the analysis of Arabic, for example, a single consonant pair would be either acceptable or unacceptable in all word forms (Berkley 1994b, Pierrehumbert 1999, Plenat 1996). Under versions of OT that permit stochastic constraint ranking, such as Boersma and Hayes (2001), probabilistic variation in the outcomes would result in probabilistic variation for all words, not in differential probabilities in the lexicon. ...
... Maltese borrowings above are also suggestive of a similarity based OCP-Place constraint in both Maltese and Italian. Berkley (1994a Berkley ( , 1994b Berkley ( , 2000) presents an extensive analysis of English that reveal gradients effects of similarity and distance on consonant co-occurrence. A study of consonant co-occurrence in Thai monosyllables from a learner's dictionary (Haas 1955) has also found gradient effects of similarity and distance on consonant co-occurrence (Frisch 2000b). ...
... In order to derive consistent realization for a particular root, stochastic OT accounts require some additional meta-evaluation mechanism that can separate the instances where evaluation is conducted to determine wellformedness from the instances where evaluation is conducted to produce an output from an input. In instances where an output is produced, the constraint ranking must be limited to one of the set of rankings that allows the root to be well-formed (Berkley 1994b). But such a fixing of constraints must not be allowed to happen in cases where free variation occurs. ...
It has long been known that verbal roots containing homorganic consonant pairs are rare in Arabic, motivating the existence of an OCP-Place constraint (Obligatory Contour Principle on place of articulation) in the phonological grammar. We explore this constraint using an on-line lexicon of Arabic roots. The strength of the constraint is quantified by the ratio of the observed number of examples of each consonant pair to the number that would be statistically expected under random combination of phonemes. We show that the strength of the effect over all pairs is a gradient function of the similarity of the consonants in the pair. A similarity metric based on natural classes is developed, which solves the formal difficulties of contrastive underspecification theory while preserving the insight that contrastiveness plays an important role in perceived similarity. This metric is applied in an explicit model of the gradient OCP constraint, which achieves a better fit to the regularities and sub-regularities of the Arabic verbal lexicon than any prior approach. Lastly, we review evidence for the psychological reality of the constraint, for its existence in related forms in other languages, and for its cognitive/phonetic foundations in the speech processing system. We argue that the total body of evidence supports a model in which phonetic and cognitive pressures incrementally affect the lexicon, and phonotactic constraints are abstractions over the lexicon of phonological forms.
... The case in question relates to proximate repetition of similar or identical phonological units within a lexical item. Most languages impose restrictions on word-internal co-occurrences of similar elements (Berkley, 1994;McCarthy, 1986;Pozdniakov & Segerer, 2007;Suzuki, 1998). 1 For example, the first two consonants in Arabic and Hebrew verb roots cannot be identical or homorganic in place of articulation (e.g., *m_m…, *b_m…, *g_k…) (Greenberg, 1950). In English, place-sharing consonants can co-occur in proximity (e.g., bib, mop, pop), but such sequences are statistically less frequent than would be expected (Berkley, 1994; see also Monaghan & Zuidema, 2015 for confirmation of the same pattern in Dutch, French and German, and Pozdniakov & Segerer, 2007 in languages across 15 families and isolates). ...
... Most languages impose restrictions on word-internal co-occurrences of similar elements (Berkley, 1994;McCarthy, 1986;Pozdniakov & Segerer, 2007;Suzuki, 1998). 1 For example, the first two consonants in Arabic and Hebrew verb roots cannot be identical or homorganic in place of articulation (e.g., *m_m…, *b_m…, *g_k…) (Greenberg, 1950). In English, place-sharing consonants can co-occur in proximity (e.g., bib, mop, pop), but such sequences are statistically less frequent than would be expected (Berkley, 1994; see also Monaghan & Zuidema, 2015 for confirmation of the same pattern in Dutch, French and German, and Pozdniakov & Segerer, 2007 in languages across 15 families and isolates). This avoidance of word-internal repetition is subject to effects of similarity and proximity; stronger co-occurrence avoidance between two sounds is observed when they are more similar in type and closer in distance. ...
The idea that natural language is shaped by biases in learning plays a key role in our understanding of how human language is structured, but its corollary that there should be a correspondence between typological generalisations and ease of acquisition is not always supported. For example, natural languages tend to avoid close repetitions of consonants within a word, but developmental evidence suggests that, if anything, words containing sound repetitions are more, not less, likely to be acquired than those without. In this study, we use word-internal repetition as a test case to provide a cultural evolutionary explanation of when and how learning biases impact on language design. Two artificial language experiments showed that adult speakers possess a bias for both consonant and vowel repetitions when learning novel words, but the effects of this bias were observable in language transmission only when there was a relatively high learning pressure on the lexicon. Based on these results, we argue that whether the design of a language reflects biases in learning depends on the relative strength of pressures from learnability and communication efficiency exerted on the linguistic system during cultural transmission.
... Many studies have shown that speakers of Semitic languages generalize the dislike of AAB stems to novel forms (e.g. Berkley 1994, Berent & Shimron 1997, Buckley 1997, Berent, Everett & Shoimron 2001, Frisch & Zawaydeh 2001. In fact, speakers demonstrably project such generalizations across the board, even to novel instances with novel phonological elements (segments, features) that are unattested in their language (Berent et al. 2002). ...
... The restrictions on phonological identity arguably exist in every language (Suzuki 1998, Walter 2007) and they are amply documented experimentally (Berkley 1994, Berent & Shimron 1997, Buckley 1997). However, not all languages exhibit morphological reduplication. ...
Does knowledge of language transfer spontaneously across language modalities? For example, do English speakers, who have had no command of a sign language, spontaneously project grammatical constraints from English to linguistic signs? Here, we address this question by examining the constraints on doubling. We first demonstrate that doubling (e.g. panana ; generally: ABB) is amenable to two conflicting parses (identity vs. reduplication), depending on the level of analysis (phonology vs. morphology). We next show that speakers with no command of a sign language spontaneously project these two parses to novel ABB signs in American Sign Language. Moreover, the chosen parse (for signs) is constrained by the morphology of spoken language. Hebrew speakers can project the morphological parse when doubling indicates diminution, but English speakers only do so when doubling indicates plurality, in line with the distinct morphological properties of their spoken languages. These observations suggest that doubling in speech and signs is constrained by a common set of linguistic principles that are algebraic, amodal and abstract.
... However, adjacency is also involved in many dissimilatory processes and cooccurrence restrictions, which do not involve feature spreading. Berkley (1994), Frisch, Broe and Pierrehumbert (to appear) document how increased distance can impact the severity of cooccurrence restrictions. I propose the following PROXIMITY constraint which constrains the distance of corresponding elements: ...
... 25. An alternative to assessing multiple violations is to have a fixed hierarchy of PROXIMITY constraints arranged to express increasing distances (Berkley 1994). Suzuki (1998) parameterizes OCP-style constraints according to such a hierarchy. ...
This article addresses the problem of coronal palatalization in Harari (Ethiopian Semitic) triggered by the 2nd person singular feminine non-perfective subject suffix/-i/. The palatalization process is unusual in two respects: (a) palatalization operates at a distance over other vowels and consonants and (b) palatalization may optionally affect more than one coronal consonant in the same stem, including prefixes. Although long-distance palatalization has been documented for other languages, it has been analyzed either as a floating affix or as consonant harmony. While Harari palatalization shares properties with both of these phenomena, it should be analyzed as neither. Harari palatalization targets a specific group of consonants and is modeled using a correspondence agreement constraint rather than aligning or spreading the palatalizing feature. This accounts for its ability to skip over intervening consonants and vowels, including front vowels and palato-alveolar consonants.
... That is, the net result of adjudicating constraint violations is the same for all input forms that share the relevant particulars. In the analysis of Arabic, for example, a single con-sonant pair would be either acceptable or unacceptable in all word forms (Berkley 1994b;Pierrehumbert 1999;Plenat 1996). Under versions of OT that permit stochastic constraint ranking, such as Boersma and Hayes (2001), probabilistic variation in the outcomes would result in probabilistic variation for all words, not in differential probabilities in the lexicon. ...
... In order to derive consistent realization for a particular root, stochastic OT accounts require some additional meta-evaluation mechanism that can separate the instances where evaluation is conducted to determine well-formedness from the instances where evaluation is conducted to produce an output from an input. In instances where an output is produced, the constraint ranking must be limited to one of the set of rankings that allows the root to be well-formed (Berkley 1994b). But such a fixing of constraints must not be allowed to happen in cases where free variation occurs. ...
It has long been known that verbal roots containing homorganic consonant pairs are rare in Arabic, motivating the existence of an OCP-Place constraint (Obligatory Contour Principle on place of articulation) in the phonological grammar. We explore this constraint using an on-line lexicon of Arabic roots. The strength of the constraint is quantified by the ratio of the observed number of examples of each consonant pair to the number that would be statistically expected under random combination of phonemes. We show that the strength of the effect over all pairs is a gradient function of the similarity of the consonants in the pair. A similarity metric based on natural classes is developed, which solves the formal difficulties of contrastive underspecification theory while preserving the insight that contrastiveness plays an important role in perceived similarity. This metric is applied in an explicit model of the gradient OCP constraint, which achieves a better fit to the regularities and sub-regularities of the Arabic verbal lexicon than any prior approach. Lastly, we review evidence for the psychological reality of the constraint, for its existence in related forms in other languages, and for its cognitive/phonetic foundations in the speech processing system. We argue that the total body of evidence supports a model in which phonetic and cognitive pressures incrementally affect the lexicon, and phonotactic constraints are abstractions over the lexicon ofphonological forms. 0.
... Slovenian is a case of restrictions specific to place of articulation. This resembles OCP restrictions in Russian (Padgett 1992), English (Berkley 1994), Arabic (Pierrehumbert 1993), and Japanese (Kawahara et al. 2006). Like the Slovenian case, these languages exhibit a dispreference (but not categorical illformedness) for consonants with the same place of articulation within a word. ...
This paper shows that a postalveolar co-occurrence restriction (Obligatory Contour Principle, OCP) is a productive component of Slovenian phonology. We first examine whether an apparent OCP-based restriction on derived palatalization, previously observed in corpus data (Jurgec 2016), extends to novel forms via a goodness-rating task. We then explore the generality of the restriction across the lexicon, in non-derived novel words as well as derived forms. Our results confirm that native speakers judge derived palatalized nonce forms to be less acceptable when the stem contains another postalveolar, reflecting the pattern found in the previous corpus study. We further demonstrate that multiple postalveolars are dispreferred even in non-derived words, which suggests that the effect is a general case of OCP. This is additionally supported by effects of proximity (the restriction is stronger for postalveolars separated only by a single vowel than for those further apart from one another) and identity (the restriction is stronger for identical than non-identical postalveolars), reflecting cross-linguistic tendencies in the manifestation of OCP and non-local consonant dissimilation. Finally, we show that the restriction does not appear to apply to all places of articulation, suggesting that the co-occurrence restriction in Slovenian specifically targets postalveolars, and adding a previously unattested pattern to the typology of OCP phenomena on consonant place.
... If this is the correct interpretation, the coordination pattern underlying both acoustic outputs[Alternatively, assume that avoidance of overlap truly weakens as the CC profile changes from equal to unequal sonority. This is consistent with work on OCP effects which does indeed suggest that a more refined notion of 'identity' is involved (Padgett 1992, Pierrehumbert 1993, Selkirk 1993, Berkley 1994). In a sequence of segments, identity of their oral gestures would be a necessary but not a sufficient condition for violating the OCP. ...
Linguistic form is expressed in space, as articulators effectconstrictions at various points in the vocal tract, but also in time, as articulators move. A rather widespread assumption in theories of phonology and phonetics is that the temporal dimension of speech is largely irrelevant to the description and explanation of the higher-level or more qualitative aspects of sound patterns. The argument is presented that any theory of phonology must include a notion of temporal coordination of gestures. Linguistic grammars are constructed in part out of this temporal substance. Language-particular sound patterns are in part patterns of temporal coordination among gestures.1
... and Bengali (Khan to appear), to take but a few examples. Berkley (1994Berkley ( , 2000 found that the same is true of English: words like king or mop, in which two sounds with the same place of articulation are separated by a vowel, are underrepresented when compared to words with heterorganic consonants like sing and mat. ...
... In spite of this, strong OCP-PLACE effects are found across many languages; for example, they are found in Semitic languages such as Arabic (e.g., Frisch, et al., 2004;Greenberg, 1950;McCarthy, 1985) and Hebrew (e.g., Berent & Shimron, 1997;McCarthy, 1985). Furthermore, many genetically and geographically unrelated languages 4 OCP-PLACE IN SPEECH SEGMENTATION display non-categorical, gradient effects of OCP-PLACE (e.g., English: Berkley, 1994;Muna: Coetzee & Pater, 2008;Niger-Congo languages: Pozdniakov & Segerer, 2007; Dutch: . In the latter type of language, pairs of consonants with shared place features are attested, such as /pVm/ in English spam or /mVb/ in mob. ...
OCP-PLACE, a cross-linguistically well-attested constraint against pairs of consonants with shared [place], is psychologically real. Studies have shown that the processing of words violating OCP-PLACE is inhibited. Functionalists assume that OCP arises as a consequence of low-level perception: a consonant following another with the same [place] cannot be faithfully perceived as an independent unit. If functionalist theories were correct, then lexical access would be inhibited if two homorganic consonants conjoin at word boundaries—a problem that can only be solved with lexical feedback.
Here, we experimentally challenge the functional account by showing that OCP-PLACE can be used as a speech segmentation cue during pre-lexical processing without lexical feedback, and that the use relates to distributions in the input.
In Experiment 1, native listeners of Dutch located word boundaries between two labials when segmenting an artificial language. This indicates a use of OCP-LABIAL as a segmentation cue, implying a full perception of both labials. Experiment 2 shows that segmentation performance cannot solely be explained by well-formedness intuitions. Experiment 3 shows that knowledge of OCP-PLACE depends on language-specific input: in Dutch, co-occurrences of labials are under-represented, but co-occurrences of coronals are not. Accordingly, Dutch listeners fail to use OCP-CORONAL for segmentation.
... We do not adopt this approach, preferring to let the models learn from all of the available data and imposing the penalty for lack of restrictiveness explicitly. Berkley 1994aBerkley , 2000Buckley 1997;MacEachern 1999;Coetzee , 2008; Kawahara et al. 2006;Coon and Gallagher 2008;Ito 2007;Pozdniakov and Segerer 2007;Walter 2007;Anttila 2008;Dmitrieva and Anttila 2008;Kager et al. 2008;Graff and Jaeger 2009). 3 McCarthy (1988,1994) links the restriction on identical place features to a more general restriction on co-occurrence of identical elements, the Obligatory Contour Principle (OCP; Leben 1973;Goldsmith 1976;McCarthy 1979McCarthy , 1981McCarthy , 1986, therefore the constraint is generally referred to as OCP-Place. ...
Subsidiary features modulate the degree to which the Obligatory Contour Principle on place (OCP-Place) is violated by homorganic consonants. Statistical analysis of consonant co-occurrence patterns in four unrelated languages, combined with a method of model compar-ison that incorporates Occam's Razor, support a theory in which each place and subsidiary feature has a weight that contributes to gradient OCP-Place violation. Crucially, the weights of subsidiary features are not free to vary across place of articulation within a language. This weighted feature theory is more restrictive than the alternative proposed by Coetzee and Pater 2008b, which allows subsidiary features to vary in weight both across languages and across places within a language (an approach that has its roots in the non-quantitative theories of Yip 1989; Padgett 1991, 1995; McCarthy 1994). The weighted feature theory is less restrictive than the natural classes model of Frisch et al. 2004, but the greater descriptive freedom pro-vided by language-particular weighting is motivated by the data even when Occam's Razor is taken into account. In addition to arguing for a novel theory of subsidiary features, this paper demonstrates that the method of evaluating analyses of co-occurrence data with respect to Ob-served/Expected (O/E) values — as originally proposed by Pierrehumbert 1993 and adopted by several recent papers in NLLT (Frisch et al. 2004, Coetzee and Pater 2008b, Anttila 2008, Coon and Gallagher 2008) — is mathematically flawed, as it confounds co-occurrence restric-tions with positional probabilities. The main empirical claim of Coetzee and Pater 2008b, namely that correlations with the data of Arabic and Muna uniformly favor their weighted-constraint theory over that of Frisch et al. 2004, is based on the problematic O/E method and must be qualified. An alternative, statistically sound method of model evaluation and com-parison based on probability theory is introduced and shown to support the weighted feature theory of OCP-Place over the alternatives considered. the participants of the Linguistics colloquium at Yale and the the CLSP seminar at Johns Hopkins for helpful feedback. We are grateful to Andries Coetzee and Stefan Frisch for making data from Muna and Arabic available to us. Bruce Hayes deserves special acknowledgment for his role in the development of the grammatical frame-work that we adopt, as well as for making versions of the Shona and Wargamay data publicly accessible.
... The second kind of evidence is not about possible and impossible words, but rather about statistical tendencies in the English lexicon. Berkley (1994Berkley ( , 2000 counted the number of English words with two homorganic consonants separated by at most two segments (i.e. pop, palm, king, skulk, state, tact, etc.). ...
In this paper, I discuss the results of word-likeness rating experiments with Hebrew and English speakers that show that language users use their grammar in a categorical and a gradient manner. In word-likeness rating tasks, subjects make the categorical distinction between grammatical and ungrammatical – they assign all grammatical forms equally high ratings and all ungrammatical forms equally low ratings. However, in comparative word-likeness tasks, subjects are forced to make distinctions between different grammatical or ungrammatical forms. In these experiments, they make finer gradient well-formedness distinctions. This poses a challenge on the one hand to standard derivational models of generative grammar, which can easily account for the categorical distinction between grammatical and ungrammatical, but have more difficulty with the gradient well-formedness distinctions. It also challenges models in which the categorical distinction between grammatical and ungrammatical does not exist, but in which an ungrammatical form is simply a form with very low probability. I show that the inherent comparative character of an OT grammar enables it to model both kinds of behaviors in a straightforward manner.
... Our study thus shows that such cooccurrence restrictions are general tendencies rather than categorical prohibitions (see, e.g. Berkley 1994; see also Moreton et al. 1998 for other aspects of phonological tendencies in the Japanese lexicon). How to capture such tendencies in the current framework of Optimality Theory (Prince and Smolensky 1993) remains an interesting theoretical question. ...
... In many languages, this constraint takes probabilistic effects: In English, for example, sequences such as /p-m/ in the word spam are not forbidden. They are just not very frequent (e.g., Berkley, 1994, see Chapter 3 and 4 for more detail). There is evidence that native Hebrew, Arabic, English and Dutch listeners have knowledge of OCP-PLACE (Berent & Shimron, 1997; Coetzee, 2008; Frisch & Zawaydeh, 2001) and use their knowledge for lexical processing (Berent, Everett, & Shimron, 2001; Kager & Shatzman, 2007, see Chapter 3 and 4 for more detail). ...
One of the most challenging tasks for language-learning infants and second language (L2)-learning adults is to segment the continuous stream of speech that surrounds them, and, following this, to acquire a lexicon. Both speech segmentation and lexical acquisition are known to be facilitated by phonotactics, i.e., language-specific restrictions on how phonemes may combine. This dissertation addresses questions regarding the representation and acquisition of such phonotactic knowledge in a native language and an L2.
Five experimental studies are presented. The first three studies, using the artificial language learning paradigm, reveal that segmentation is influenced by structural phonotactic knowledge of OCP-PLACE, a restriction against pairs of consonants sharing the feature [Place]. It is shown that this knowledge is used only by native listeners or advanced L2 learners of a language restricted by the constraint. This suggests a language-specific acquisition from the input. The third study, with infants, shows that this input is continuous speech rather than the lexicon. The remaining two studies demonstrate that abstract phonotactic knowledge of syllable structure is represented separately from specific probabilistic knowledge, as the two have separate effects on lexical acquisition in a short-term memory recall task. Moreover, results from L2 learners suggest that probabilistic knowledge can be acquired independently of structural knowledge of the L2.
While most studies have looked at the influence of specific representations of phonotactic probability, here it is shown that representations of abstract structural constraints also influence processing. Moreover, it is demonstrated that both types of phonotactic representations are acquired from the input
... First, recent research has shown that the Obligatory Contour Principle (OCP) emerges gradiently in phonotactics. In particular, Berkley (1994) demonstrated a gradient OCP-place effect in the English lexicon: monomorphemic monosyllabic words where the onset and the coda share place of articulation are statistically underrepresented. We replicate and extend Berkley's result using CELEX2 (Baayen, Piepenbrock, & Gulikers, 1995) and CMU (Weide, 1998) ...
... The measure is a general statistical notion, and has recently been deployed in linguistic work that analyzes combinability of two linguistic elements in corpora (e.g. Berkley, 1994;Coetzee and Pater, 2008;Frisch, 2000;Frisch et al., 2004;Kawahara, 2007;Kawahara and Shinohara, 2009;Pierrehumbert, 1993;Shatzman and Kager, 2007). O/E ratios are ratios between O-values (the actual occurrences of the pairs observed in the corpus) and E-values (how often the pairs are expected to occur if their two individual elements are combined randomly). ...
... This grouping is based on four major place classes, but the coronal class is further divided into two subclasses. This classification follows the previous studies on consonant co-occurrence patterns, which have shown that coronal sonorants and coronal obstruents constitute separate classes, as in Arabic (Frisch et al. 2004;Greenberg 1950;McCarthy 1986McCarthy , 1988, English (Berkley 1994), Javanese , Russian (Padgett 1992), Wintu (McGarrity 1999), and most importantly in this context, Japanese . The rhyming patterns of the five classes are summarized in Table 5. ...
Using data from a large-scale corpus, this paper establishes the claim that in Japanese rap rhymes, the degree of similarity of two consonants positively correlates with their likelihood of making a rhyme pair. For example, similar consonant pairs like {m-n}, {t-s}, and {r-n} frequently rhyme whereas dissimilar consonant pairs like {m-∫}, {w-k}, and {n-p} rarely do. The current study adds to a body of literature that suggests that similarity plays a fundamental role in half rhyme formation (A. Holtman, 1996, A generative theory of rhyme: An optimality approach, PhD dissertation. Utrecht Institute of Linguistics; R. Jakobson, 1960, Linguistics and poetics: Language in literature, Harvard University Press, Cambridge; D. Steriade, 2003, Knowledge of similarity and narrow lexical override, Proceedings of Berkeley Linguistics Society, 29, 583–598; A. Zwicky, 1976, This rock-and-roll has got to stop: Junior’s head is hard as a rock. Proceedings of Chicago Linguistics Society, 12, 676–697). Furthermore, it is shown that Japanese speakers take acoustic details into account when they compose rap rhymes. This study thus supports the claim that speakers possess rich knowledge of psychoacoustic similarity (D. Steriade, 2001a, Directional asymmetries in place assimilation. In E. Hume, & K. Johnson (Eds.),
The role of speech perception in phonology (pp. 219–250). San Diego: Academic
Press.; D. Steriade, 2001b, The phonology of perceptibility effects: The P-map and its
consequences for constraint organization, ms., University of California, Los Angeles;
D. Steriade, 2003, Knowledge of similarity and narrow lexical override, Proceedings
of Berkeley Linguistics Society, 29, 583–598).
... As we pointed out earlier, however, this evidence is consistent with either root-or stem-based accounts. Indeed, recent linguistic (e.g., Berkley, 1994;Kawahara, Hajime, & Kiyoshi, 2006) and psycholinguistic evidence (Bonatti et al., 2005;Newport et al., 2004) suggests that the restriction on consonant co-occurrence across intermediate vowels is not unique to Semitic. Thus, although adult speakers (of both Semitic and non-Semitic languages) might well detect relationships between 11 The fact that the Piú l and Péel paradigms differ on their potential to violate the OCP does not demonstrate that this factor is the cause of our experimental results. ...
Is the structure of lexical representations universal, or do languages vary in the fundamental ways in which they represent lexical information? Here, we consider a touchstone case: whether Semitic languages require a special morpheme, the consonantal root. In so doing, we explore a well-known constraint on the location of identical consonants that has often been used as motivation for root representations in Semitic languages: Identical consonants frequently occur at the end of putative roots (e.g., skk), but rarely occur in their beginning (e.g., ssk). Although this restriction has traditionally been stated over roots, an alternative account could be stated over stems, a representational entity that is found more widely across the world's languages. To test this possibility, we investigate the acceptability of a single set of roots, manifesting identity initially, finally or not at all (e.g., ssk versus skk versus rmk) across two nominal paradigms: CéCeC (a paradigm in which identical consonants are rare) and CiCúC (a paradigm in which identical consonants are frequent). If Semitic lexical representations consist of roots only, then similar restrictions on consonant co-occurrence should be observed in the two paradigms. Conversely, if speakers store stems, then the restriction on consonant co-occurrence might be modulated by the properties of the nominal paradigm (be it by means of statistical properties or their grammatical sources). Findings from rating and lexical decision experiments with both visual and auditory stimuli support the stem hypothesis: compared to controls (e.g., rmk), forms with identical consonants (e.g., ssk, skk) are less acceptable in the CéCeC than in the CiCúC paradigm. Although our results do not falsify root-based accounts, they strongly raise the possibility that stems could account for the observed restriction on consonantal identity. As such, our results raise fresh challenge to the notion that different languages require distinct sets of representational resources.
Phonological regressions or U‐shaped development have frequently been observed in longitudinal studies of child speech production. However, the typology of which phonological patterns regress, and their implications for learning, have not been given much attention in the recent literature on constraint‐based phonological development. One basic question is simply the definition of a phonological regression, as created by the grammar or other mechanisms, which is in turn dependent on the type of grammar and learner assumed. This paper systematically addresses the question of whether or not attested phonological regressions are incompatible with an error‐driven approach to grammatical development, whereby each round of learning is predicted to move the learner closer to the target language. From this perspective, this survey discusses case studies of phonological regression in the literature, grouped according to their ease of explanation under error‐driven learning. Three types are identified and exemplified: regressions which are easily explained with existing error‐driven algorithms for constraint‐reranking; regressions which can also be derived through error‐driven learning by adopting an additional tool (for incorporating child‐specific phonetic experience); and regressions whose error‐driven motivation remains unclear. Another central theme of the survey is the degree of variation and lexical exceptionality among these regression patterns, and the extent to which such variability is captured in the learner's algorithms or grammar. Interim conclusions are provided, and necessary future directions for empirical and theoretical research are discussed.
This article is categorized under:
• Linguistics > Language Acquisition
• Linguistics > Linguistic Theory
Duality of patterning, is, by hypothesis, a universal design feature of language. Every language constructs words from meaningful units (morphemes), which, in turn, are comprised of meaningless phonological elements (e.g., segments, syllables). But whether the language faculty does, in fact, include a separate morphological level, distinct from the phonology, is a matter of controversy. To elucidate the role of morphology, here we ask whether morphological forms are constrained by putatively universal combinatorial principles, distinct from those applying to phonological patterns. Our research exploits the structural ambiguity of doubling. Doubling (e.g., trafraf) is open to two competing interpretations-as either a purely phonological form, or as a complex morphological structure that is systematically linked to meaning (e.g., trafraf is the diminutive of traf). Our experiments show that responses to doubling (trafraf) shift radically, depending on its level of analysis. Viewed as a meaningless phonological form, doubling is dispreferred irrespective of its kind (i.e., trafraf is as bad as traftaf, even though the latter violates a morphological constraint on contiguity). But once doubling is systematically linked to meaning (i.e., as a morphological structure), the doubling dislike shifts into a reliable preference, and an additional constraint on its contiguity arises (i.e., trafraf>traftaf). Remarkably, the dissociation between morphological and phonological doubling emerges regardless of whether morphological reduplication is abundant in participants' language (in Hebrew) or relatively rare (in English). These results suggest the existence of distinct linguistic constraints that preferentially target the morphological vs. phonological levels. We discuss various explanations for the origins of these restrictions.
In this paper, I provide a unified account of three frequency effects in phonology. First, typologically marked elements are underrepresented. Second, phonological changes are underrepresented. Third, morphologically conditioned phonological changes are overrepresented. These effects are demonstrated with corpus data from English and Welsh. I show how all three effects follow from a simple conception of phonological complexity. Further, I demonstrate how this notion of complexity makes predictions about other phenomena in these languages, and that these predictions are borne out. I model this with traditional Optimality Theory, but the proposal is consistent with any constraint-based formalism that weights constraints in some way.
This study examines vowel epenthesis in Hebrew verbs with stem-medial glottals. The stem-internal epenthetic vowels of these verbs colloquially display variation in 3rd person feminine forms in some verb templates. While the normative epenthetic vowel in these cases is a, there are cases where it is colloquially e. We provide empirical evidence for the variation (or lack thereof), accounting for the differences 2 among verb templates and the variation by appealing to paradigmatic faithfulness constraints, and grammatical components deriving verbs. We argue that vowel selection is motivated by competing faithfulness to three different paradigms: other feminine 3rd person forms, base forms, and the general inflectional paradigm. The component of the grammar deriving verbs, lexicon vs. syntax, as well as the way verbs are stored in the lexicon, determine which paradigmatic relation is relevant in different cases. The results of this study show the importance of paradigm accessibility in morpho-phonological processes. In addition, they point to a high correlation between the degree of variation in such processes and the locus of application and storage.
Highlights
• If OCP-Labial holds as a gradient constraint, specific labial pairs can be exempt.
• Dutch listeners know of such exceptions, which affect their processing of speech.
• Phonotactic knowledge influenced their segmentation of artificial languages.
• Detailed phonotactic knowledge affects processing when task demands are simple.
• Abstract phonotactic knowledge may affect processing when task demands are complex.
Abstract
Many languages restrict their lexicons by OCP-Place, a phonotactic constraint against co-occurrences of consonants with shared [place] (e.g., McCarthy, 1986). While many previous studies have suggested that listeners have knowledge of OCP-Place and use this for speech processing, it is less clear whether they make reference to an abstract representation of this constraint. In Dutch, OCP-Place gradiently restricts non-adjacent consonant co-occurrences in the lexicon. Focusing on labial-vowel-labial co-occurrences, we found that there are, however, exceptions from the general effect of OCP-Labial: (A) co-occurrences of identical labials are systematically less restricted than co-occurrences of homorganic labials, and (B) some specific pairs (e.g., /pVp/, /bVv/) occur more often than expected. Setting out to study whether exceptions such as (A) and (B) had an effect on processing, the current study presents an artificial language learning experiment and a reanalysis of Boll-Avetisyan and Kager's (2014) speech segmentation data. Results indicate that Dutch listeners can use both knowledge of phonotactic detail and an abstract constraint OCP-Labial as a cue for speech segmentation. We suggest that whether detailed or abstract representations are drawn on depends on the complexity of processing demands.
The internal structure of the syllable has been a matter of long-standing debate. Some theories propose a highly articulated tree structure whose topmost constituents are the onset and the rhyme. However, much recent work in phonological theory has adopted a flatter model in which neither the onset nor the rhyme constitutes a constituent.
Experiments using the novel word game paradigm developed in Treiman (1983) have been interpreted as supporting the onset-rhyme model. Here we present new results using this paradigm. Subjects learned to insert infixes into simple monosyllabic words, and then extended the infixation to more complex forms including longer words with variable stress placement. The results are interpreted in the light of findings about morphophonemic processes occurring in natural language. Our model draws on the concept of template mapping, adapted from the literature on prosodic morphology. The patterns observed in the data are better modeled by mapping onto output templates than by any derivational rule referring to onset-rhyme constituency. Though the output templates do include prosodic detail, the level of detail available in flat models of the syllable is sufficient to explain the results. A critical appraisal of these results in relation to other results in the literature leads us to reject the onsetrhyme model.
This thesis investigates the factors influencing the adaptation of foreign words to
English, beyond traditional phonological parameters such as sonority distance. The
data examined were produced in an on-line adaptation task to study purely linguistic
rather than orthographic or historical influences. The adapted words contain
only lesser-studied phonotactic problems rather than segmental ill-formedness. The
choice of Russian as a donor and English as a borrowing language allow the study
of adaptations in a setting which allows a further strategy of alteration of ill-formed
consonant clusters beyond vowel epenthesis and consonant deletion, namely the
substitution of segments to change one cluster into another. In contrast to previous
research, English production of Russian stimuli with initial consonant clusters
showed that segment change is applied frequently, comparable to the amount of
vowel epenthesis. Extensive variation was observed, both in ratio of successful
production, and in the choice and distribution of adaptation strategy.
The factors in adaptation investigated were the sonority distance of the foreign
clusters, as well as concepts which have received much recent attention within
phonology, namely gradient grammaticality, similarity and frequency: English native
speaker judgments were collected about the perceived grammaticality of foreign
clusters and the similarity between targets and adaptations, while the frequency of
possible adaptations in English was calculated from a corpus of spoken English. Results
show that sonority cannot explain the variation in adaptation. Furthermore,
frequency has no influence on the choice of adaptation; however, higher perceived
badness results in a higher percentage of adaptations, and perceived similarity is
decisive for the choice of adaptations. A comparison of similarity judgments of English
and Russian listeners suggested that, in keeping with Steriade (2001), there
are some cross-linguistically corresponding rankings of similarity; however, differences
between languages due to phonotactics and phonetic detail were also found.
In summary, the experiment results suggest that the adaptation of loanwords occurs
in both in perception and production; furthermore, it is determined both by
L1 specifics and cross-linguistic tendencies, an thus neither a straightforward application
of L1 phonology nor completely independent of language background.
The psychological reality of an abstract consonant dissimilation constraint is demonstrated in an experiment with native speakers of Jordanian Arabic. In this experiment, novel verbs containing constraint violations and those without violations were presented orthographically for judgments of well-formedness. Native speaker well-formedness judgments reflected knowledge of the phonotactic constraint. Systematic gaps were rated much less wordlike than accidental gaps that were equivalent in their lexical characteristics. Judgments for novel verbs containing constraint violations were also gradiently influenced by consonant pair similarity. The experimental study supports previous dictionary-based phonotactic analyses that propose that the native speaker's knowledge of consonant cooccurrence constraints in Arabic is based on emergent generalizations over the lexical items in an abstract root lexicon.
I present evidence from Navajo and English that weaker, gradient versions of morpheme-internal phonotactic constraints, such as the ban on geminate consonants in English, hold even across prosodic word boundaries. I argue that these lexical biases are the result of a maximum entropy phonotactic learning algorithm that maximizes the probability of the learning data, but that also contains a smoothing term that penalizes complex grammars. When this learner attempts to construct a grammar in which some constraints are blind to morphological structure, it underpredicts the frequency of compounds that violate a morpheme-internal phonotactic. I further show how, over time, this learning bias could plausibly lead to the lexical biases seen in Navajo and English.*
Acquisition and the lexicon collects revised versions of nineteen primary papers and four commentaries presented at the 5th Conference on Laboratory Phonology, held at Northwestern University in July 1996. It is not possible to do justice to each paper in a collection this large and diverse, so individual papers will be discussed as they relate to three themes that emerge from the book as a whole: (i) the relationship between phonology and speech processing – this volume contains more papers on speech processing than its predecessors in the Laboratory Phonology series, which brings to the forefront questions about the relationship between phonological theory and theories of speech processing; (ii) the phonetics–phonology divide – a number of the phonology papers question the common assumption that phonetics and phonology are separate components of grammar; (iii) frequency effects – results are presented showing effects of the frequency of words and other phonological units in a striking variety of domains.
In this paper, I present an analysis of the typology of laryngeal co-occurrence restrictions based on contrast markedness. The key ingredient of the analysis, for which I provide experimental support, is that laryngeal co-occurrence phenomena reflect a preference for maximising the perceptual distinctness of contrasts between words (Flemming 1995, 2004). An AX discrimination task finds that the contrast between an ejective and a plain stop is less accurately perceived in the context of another ejective in the word than in the context of another plain stop in the word. Pairs of words like [k'ap'i] and [k'api], which contrast 2 vs. 1 ejectives, are less reliably distinguished than pairs of words like [kap'i] and [kapi], which contrast 1 vs. 0 ejectives. The unifying factor of all laryngeal co-occurrence patterns is the neutralisation of the contrast between words with one and two laryngeally marked segments, exactly the contrast that is shown to be relatively perceptually weak.
Building on the analysis of long-distance geminates as reduplication, this article argues that the OCP may apply to identical consonants across an intervening vowel. This is adduced from the behavior of guttural consonants in Semitic. It is further argued that antigemination is not resistance to an OCP violation but avoidance of gemination; syncope between identical consonants avoids an OCP violation by creating a geminate. This entails that there is no surface representa-tional distinction between true and fake geminates. Finally, cases of reduplication are examined in which the standard reduplicant is changed to avoid either gemination or an OCP violation.
1 The rationale This volume "Papers on Japanese puns" is a collection of papers on Japanese imperfect puns that I wrote between 2007 and 2009 (often in collaboration with Professor Kazuko Shinohara, the Tokyo University of Agriculture and Technology), although most of the papers appear in print in 2009 and 2010. There are several reasons to put together a volume like this. First, I believe that I have done a wide range of analyses on Japanese imperfect puns, and it is convenient to have a collection of relevant papers. Second, although as much as I liked—and still do like—the project, I feel that I have done enough and am ready to move on to other projects. This collection would thus serve as a closure on my side (though I plan to build my future research partially on this project). Third, related to the first two reasons, I hope that building on these works, somebody interested in this project can take up some remaining issues, and that this volume serves as a useful reference. 1 2 A brief history Here is a somewhat personal history of how the research was developed. I was first inspired by a colloquium talk by Professor Donca Steriade at the University of Massachusetts in 2003, and started working on Japanese rap rhyming, the result of which is included as the final paper of this volume. Professor Kazuko Shinohara and I applied the same methodology to Japanese imperfect puns, and wrote the third paper in this volume. Then, we decided that imperfect puns has lots of potential, providing a rich resource for linguistic analyses. In developing this research program, we also realized its pedagogical values because of its accessibility. For example, the sixth paper included in this volume was based on a BA thesis study by Nobuhiro Yoshida at the Tokyo University of Agriculture and Technology under the supervision of Professor Shinohara. Two of my undergraduate students at Rutgers, Ayanna Beatie and Allan Schwade, worked on Korean tongue twisters from a similar perspective and they presented their work at the undergraduate linguistic conferences at Harvard. Moreover, the materials attract the interest of students in introductory classes, and also the interest of audience outside the field—I have been asked to give my talk on Japanese rap rhymes and puns to various non-professional audiences. 1 Interested readers are also referred to my website for further information as well, which I will try to update frequently.
Past theoretical analyses have claimed that some languages employ a special type of phonological spreading of a consonant over a vowel, long-distance consonantal spreading. I argue that this type of spreading can and must be eliminated from the theory, by reducing it to segmental copying as in reduplication. This elimination is first motivated from a number of perspectives, including considerations of locality and theoretical redundancy. The reduction to reduplication is then developed in detail for Temiar, one of the main indigenous languages of Malaysia, notorious for the complexity of its copying patterns. Crucial to this reduction is the notion of gradient violation of constraints in Optimality Theory (Prince and Smolensky 1993), and the notion of correspondence, with its particular application to reduplication (McCarthy and Prince 1995a). The proposal extends to other languages (e.g., Arabic, Chaha, Modern Hebrew, and Yoruba), where the putative spreading had been thought necessary. The elimination of long-distance consonantal spreading is argued to further obviate two other special mechanisms, also thought to apply on a language-particular basis: (a) the representation that segregates vowels and consonants on different planes, known as V/C planar segregation, and (b) the distinct mode of word formation consisting of mapping segments to templates.
Lexical items can be more or less well-formed depending on the phoneme combinations they contain. This phenomenon is called
gradient phonotactics. We propose an approach to gradient phonotactics based on Optimality Theory (Prince and Smolensky 1993/2004). At the heart of the proposal is the Complexity Hypothesis that attributes the relative well-formedness of a lexical item
to its relative grammatical complexity measured in terms of ranking information: the more complex the lexical item, the less
well-formed it is. The theory orders linguistic structures in an implicational hierarchy that reflects their relative well-formedness.
Some implications are universal; others depend on language-specific rankings. The Complexity Hypothesis is supported by phonotactic
data from Muna (Austronesian) as recently analyzed by Coetzee and Pater (2008).
How are violations of phonological constraints processed in word comprehension? The present article reports the results of an event-related potentials (ERP) study on a phonological constraint of German that disallows identical segments within a syllable or word (CC(i)VC(i)). We examined three types of monosyllabic late positive CCVC words: (a) existing words [see text], (b) wellformed novel words [see text] and component (c) illformed novel words [see text] as instances of Obligatory Contour Principle non-word (OCP) violations. Wellformed and illformed novel words evoked an N400 effect processing in comparison to existing words. In addition, illformed words produced an enhanced late posterior positivity effect compared to wellformed novel words. obligatory contour Our findings support the well-known observation that novel words evoke principle higher costs in lexical integration (reflected by N400 effects). Crucially, modulations of a late positive component (LPC) show that violations of phonological phonotactic constraints influence later stages of cognitive processing even constraints when stimuli have already been detected as non-existing. Thus, the comparison of electrophysiological effects evoked by the two types of non-existing words reveals the stages at which phonologically based structural wellformedness comes into play during word processing.
This article reports the results of speech error elicitation experiments investigating the role of two consonant co-occurrence restrictions in the productive grammar of speakers of two Ethiopian Semitic languages, Amharic and Chaha. Higher error rates were found with consonant combinations that violated co-occurrence constraints than with those that had only a high degree of shared phonological similarity or low frequency of co-occurrence. Sequences that violated two constraints had the highest error rates. The results indicate that violations of consonant co-occurrence restrictions significantly increase error rates in the productions of native speakers, thereby supporting the psychological reality of the constraints.
Introduction Increased similarity between consonants correlates with increased susceptibility to speech errors (Nooteboom 1967, MacKay 1970, Fromkin 1971, Shattuck-Hufnagel and Klatt 1979, van den Broecke and Goldstein 1980, Levitt and Healy 1985, Wilshire 1999) It follows that combinations of similar consonants may be dispreferred in languages due to production processing problems This dispreference may be phonologized as grammatical constraints (Hansson 2001 a,b, Rose & Walker 2001 ) Interesting question to establish a parallel between the kinds of consonant combinations that result in speech errors and those that are attested as grammatical co-occurrence constraints 2 Central question Does the presence of a phonological constraint on consonant co-occurrence correlate with marked production problems with those consonants? Secondary questions Do similar consonants result in more production errors than more dissimilar consonants? Do less frequent consonant combinations result i
ResearchGate has not been able to resolve any references for this publication.