Abstract:
To find out what has been and has not been accomplished in L1 and L2 acquisition research on the temporal properties of natural languages, the present review of literature does not present empirical findings of any specific languages in chronological order; rather, it synthesizes a diversifying body of empirical research across multiple L1 and L2 populations that provides hypothesis-testing, prediction-comparing results of the Primacy of Aspect Hypothesis (POA), which is composed of two subsets hypothesis, namely the association hypothesis and the developmental hypothesis. POA attracts much attention and receives both supporting and countering empirical data, with supporting data collected mostly from leaners’ production tasks and countering data from interpretation tasks. Despite the controversies, there appears to be general association patterns between lexical aspect and grammatical aspect as stated by POA, which further breeds theories to explain this phenomenon. Aimed at providing a roadmap for the many-branched POA related researches and find under-researched areas and directions for further experimental studies, this review concentrates on presenting the existing researches concerning the generative-functional debate of aspect acquisition and explaining their findings systematically, with a special focus on those using the generative approach.
Introduction
The first language (L1) and second language (L2) acquisition of aspect has been one of the most prolific topics of research in the field of language acquisition in the past few decades. This abundance of studies, on the one hand, stems from the complexity of aspect as a universal linguistic phenomenon appearing in almost all the natural languages which lies at the interface between lexicon and morphology, lexicon and syntax as well as syntax and semantics. On the other hand, its attraction lies in the fact that aspect constitutes an essential part of both language use (reflected in the lexicon and the grammar) and language acquisition (reflected in the early emergence of conceptual and linguistic expressions).The study of the acquisition of aspect can provide significant insights into the understanding of general theoretical issues in language acquisition.
The article is structured as follows: first, a comprehensive review of tense and aspect system and the possible mapping mechanisms is provided to facilitate the understanding of POA by locating it in a broader context. Then two subsets hypothesis of POA are introduced along with empirical studies testing them. To better understand the theoretical contribution of POA related studies to those of universal grammar, studies adopting the generative approach are summarized after background information of generative-functional debate is provided. Then a summary of current findings of generative studies and unanswered researched questions are presented in the conclusion.
Literature Review
1. The intricate tense and aspect system
Since a clear linguistic analysis of a particular domain serves as a good starting point for understanding the acquisition of that domain, the mechanisms of how tense and aspect work together to represent temporal properties of events are first presented here.
Although both aspect and tense are concerned with time, they are concerned with time in different ways. Tense is a deictic category, i.e. locates situations[ here the ‘situation’ is meant as a general term covering events, processes, actions, states, etc., as expressed by the verb phrase or the construction. Sometimes the distinction between states and non-states is referred to as states and actions. However, the term ‘action’ is also used in a more restricted sense, for a dynamic situation that requires the involvement of an agent. Similarly, the term ‘event’ is used to refer to a dynamic situation viewed perfectively, and the term ‘process’ - to a dynamic situation viewed imperfectively.] in time axis relative to a reference time, which is usually the speech time (Comrie,1985). When an event or state takes place or holds before the speech time, the tense is past tense; when the situation is reversed, the tense is future tense; when a process or state overlaps with the speech time, the tense is present tense. Unlike tense, which is situation-external time concerned with relating the time of the situation to a reference time, aspect is situation-internal and non-deictic. It is concerned with the perspective taken on the internal temporal organization of the situation, and so 'aspects' distinguish different ways of viewing the internal temporal constituency of the same situation (Comrie,1976). Aspectual meaning of a clause can be broken up into two separate but interactive components termed viewpoint aspect and situational aspect (Smith 1991,1997).
Viewpoint aspect, also called grammatical aspect in the literature because it is often realized through grammaticalized affixes or auxiliaries, refers to the encoding of situation time with relation to reference time. Cross-linguistically, the most basic grammatical aspectual opposition encoded is that of perfective and imperfective. As defined by Comrie (1976), perfective aspect presents a situation as an unanalyzed completed whole viewed as if from outside (external view), whereas imperfective aspect presents a situation as an ongoing, incomplete action or state, as if viewed from within (internal view).
Situational aspect, often known as lexical aspect or 'aktionsart' (plural: aktionsarten), a German word meaning ‘kinds of action’, originates from the fact that situations unfold in time in different ways. Vendler (1967) distinguishes four types of situation: state, activity, accomplishment and achievement. They differ in the temporal properties of dynamism, durativity, and telicity, as shown in the table below:
Table 1. Vendler's(1957/1967) Classification of Situations:
Situation type Dynamic Telic Punctual
State (e. g., love you, know the answer) – – –
Activity (e. g., laugh, run, walk) + – –
Accomplishment (e. g., read a book, bake a cake, paint a picture) + + –
Achievement (e. g., die, win a race, reach the summit) + + +
What worth to mention is that aspect sometime also has deictic function as tense does. For example, in a sentence like John was reading when I entered, it might seem that the different aspectual forms of the two clauses serve a deictic function of locating my entry internally to John's reading, but this apparent deictic function is only a secondary consequence of the different ways of view the internal constituency of the situations: since was reading places us internally to the reading situation, therefore naturally when we are presented with another situation given to us as a unified whole without internal constituency, this new situation is located temporally at that point in time where we already are, namely internally to John's reading.
This fact that aspect and tense can serve the same function is partly the reason that the grammaticalized expressions of aspect in languages are sometimes used also as tense marker. Many Germanic and Romance language, for instance, have conflated deictic past and aspectual meanings in their verbal morphology. For example, the English -ed marks both past tense and perfective aspect (Spanish preterite, Italian passato prossimo and French passé composé markers are also past and perfective). If we were to investigate the acquisition of aspect without the interference of tense, we need to look at languages that have separate grammatical aspect and tense morphology. Such languages include Mandarin Chinese (which has no overt tense but ample aspect marking such as the progressive marker ‘zai’ and perfective marker ‘le’), Japanese (which marks past tense and aspect separately) and all Slavic languages(which has separate aspect and tense marking). Thus, empirical studies of these languages will be paid more attention when reviewing and reporting the findings.
Note that now we have three different temporal contrasts enter play: past vs. present deictic tense, perfective vs. imperfective grammatical aspect, and lexical aspect distinctions like telic vs. atelic, stative vs. dynamic, punctual vs. durative.
Logically, there are many possible learning sequences of the three sets of variables. Some (Bronckart and Sinclair, 1973; Antinucci and Miller, 1976; Bloom et al., 1980) argued that initially lexical aspect is mapped onto tense. That is, children use past and present morphology to mark the more salient aspectual distinctions instead. In other words, children use aspect to learn tense. Olsen (1997) and Olsen and Weinberg (1999), however, consider the initial mapping to be that of lexical aspect onto grammatical aspect. Wagner (2001) argues for a third type of mapping, namely, that of grammatical aspect onto tense. These three hypotheses are illustrated below in Figure 1, with the rectangular box indicating the three variables and the arrows showing the possible mapping direction.
Figure 1. The possible mapping relationship between tense, grammatical aspect, and the Vendler’s (1967) situational aspect in language acquisition
2. The Primacy of Aspect Hypothesis
Based on the fact that grammatical aspect and tense often conflate in their forms and functions as aforementioned and that a situation might be modified by tense and aspect in terms of temporal interpretation, it is generally agrees that the syntactic structure underlying temporal interpretation of sentences are as follows:
[Tense [Grammatical Aspect* [situational aspect] ] ].
Thus, these two factors might work more closely together and more independent from lexical aspect in the acquisition process. A fourth type of mapping theory, that is, in first and second aspect acquisition, lexical aspect is mapped onto tense and grammatical aspect is proposed by opponents (Andersen & Shirai ,1996; Chen & Shirai, 2010; Jin & Hendriks,2003 and others)of the Primacy of Aspect Hypothesis(henceforth POA)), as shown in Figure 2.
Figure 2. POA proposal for tense–aspect mapping
The Primacy of Aspect model argue that lexical semantic classes guide early language learners in their production of inflectional morphology. Specifically, the POA model proposes the following four associations (Andersen & Shirai,1996, p.533) between the lexical class of the verb and grammatical marking:
1. Children first use past or perfective marking on achievement and accomplishment verbs, eventually extending its use to activity and stative verbs.
2. In languages that encode the perfective/imperfective distinction, imperfective past develops later than perfective past, and imperfective past marking begins with stative verbs and activity verbs, then extending to accomplishment and achievement verbs.
3. In languages that have progressive aspect, children first use progressive aspect marking mostly with activity verbs, then extend to accomplishment and achievement verbs.
4. Children do not incorrectly overextend progressive aspect markings to stative verbs.
The generalizations of the Primacy of Aspect Hypothesis can be reformulated into two hypothesis, the association hypothesis which says leaners initially restrict uses of perfective-past forms to telic verbs and imperfective forms to atelic verbs, and development hypothesis, and the developmental hypothesis which predict more advanced learners will exhibit a less restricted lexical aspect and tense-aspect markers association. The emergence order of combination of situation aspect with the other two variables in language production of learners predicted by the POA is illustrated in Table 2, with the arrows indicating the learning sequence.
Table 2. Order of Emergence Predicted by the Aspect Hypothesis (Shirai, 1995)
State Activity Accomplishment Achievement
Perfective-past 4← 3← 2← 1←
Imperfective-past 1→ 2→ 3→ 4→
Progressive X 1→ 2→ 3→
*X signals combination rarely occurs
POA is intended to be applied to the acquisition of tense-aspect morphology in both LI and L2 learners. Moreover, the POA model proposes that the associations outlined above will hold universally across language. Researchers have tested these associations in typologically different languages such as English, Spanish, Japanese and Chinese and have argued that their findings provide support for the claims outlined above.
By and large, there appears to be a consensus on the L1 and L2 acquisition sequences. The systematic examination of children's first language acquisition of tense and aspect begins in 1970s when numerous longitudinal studies concerned with productive data were conducted to examine children’s learning sequences and mechanisms(Brown ,1973; Bloom, Lifter, and Hafitz, 1980; Harner,1981 for English-speaking children; Bronckart and Sinclair,1973 for French; Antinucci and Miller,1976 for Italian; Aksu 1978, 1988, for Turkish; Stephany 1981, 1997 for Turkish. Li, P. and Shirai, Y. 2000 for English, Chinese and Japanese). Most of the studies cited above observed co-occurrences patterns between temporal morphology and the inherent semantics of verbs that are generally consistent with the sequences listed above, which inspired the study of L2 aspect acquisition started in mid 1980s. Andersen (1986, 1989, 1991) was among the first to generalize the emergence sequences of POA above by analyzing longitudinal, conversational data from two adolescent native English speakers learning Spanish in a naturalistic setting. Other studies range across a handful of L2s, and with participants with a diversity of L1 backgrounds reported consistent learning sequences. (Andersen, 1991; Liskin-Gasparro, 2000; Salaberry, 2002, 2003 for English-speaking L2 learners; BardoviHarlig, 1992 for L2 French learners of multiple linguistic backgrounds; BardoviHarlig&Bergstr?m, 1996 for a comparison between the acquisition of L2 English for Francophones and of L2 French for Anglophones; Cadierno, 2000 for L1 Danish learners of L2 Spanish; Collins, 2002, 2005 for L1 English speakers learning L2 French; Izquierdo& Collins, 2008 for a comparison of Anglophone and Hispanophone learners of L2 French, among others).
To summarize, results from the L1 and L2 studies provide converging evidence on the association hypothesis. However, the developmental hypothesis remains controversial in the L2 acquisition in that some researchers (Liskin Gasparro, 2000; McManus, 2013; Salaberry, 2002, 2011)find opposite developmental sequences as predicted, that is, the more advanced the learner’s level is, the more they tend to restrict their lexical and grammatical aspect correlation to the prototypical ones, directly countering Andersen’s (1991) claim that learners’ dependence on lexical aspect in the selection of verbal morphemes decreases as proficiency level increases. Forexample, Liskin-Gasparro (2000) verified that her study’s participants – advanced L2 Spanish learners consistently use the past perfective of the achievement verb ‘ver’ (to see) even in contexts where the imperfective form ‘veía’ was warranted, even after classroom instruction of Spanish and prolonged contact with native speakers. Salaberry (2011) revealed consistent results in his study of advanced Anglophone L2 Spanish learners, and McManus (2013) reached the same conclusion for advanced L2 French learners.
These studies have demonstrated that the POA accurately predicts development at lower levels of proficiency, but this hypothesis is less precise in predicting learner behavior at higher levels. If learners continue to rely on lexical aspect at high levels of acquisition, they do not expand their usage of grammatical morphemes to non-prototypical contexts, as is predicted by Andersen (1991). However, these claims do not negate evidence for a developmental sequence of aspectual morphemes; in fact, they support the notion that there exists a predictable way in which L2 learners acquire aspectual morphology. To explain the association hypothesis and to account for the failure of L2 leaners’ extension from more prototypical associations to more diversified use of grammatical aspect, this review presents studies adopting generative approach. But before that, a general account of generative and functional approach is presented in the next section to provide the necessary background information.
3.Universals in First Language Acquisition
Since POA proposes universal learning sequences across different first language learners, the question of where this universal aspect of language acquisition comes from serves as a wrestling ground for the nativist-functionalist debate, which is rooted deeply in cognitive science in the opposition between viewing the human mind as a modular system whose architecture is largely predetermined or innate (Chomsky 1988; Fodor 1983; Pinker 1994) and viewing it as a highly interactive system in which multiple components can interact simultaneously at all levels (Rumelhart &McClelland,1986; Elman et al,1996).Nativists argue that certain semantic categories such as the semantic features that define semantic categories of lexical aspect (e.g., telicity, punctuality, and dynamicity) and their ontological substrates are innately specified in the human genome and is a language universal(e.g., Bickerton 1981; Pinker 1984, 1989; Smith 1997).In contrast, functionalists adopt a connectionist, input-based, probabilistic learning approach to the problem of the acquisition of aspect, trying to account for universal aspects of first language acquisition by relying on multiple factors, such as saliency, frequency, discourse structure, information organization, processing cost, prototypical categorization, functional typology, pragmatic principle and form-function mapping(e.g., Hopper, 1979; Longacre, 1981;DeLancey 1998;Elman et al., 1996;van Hout and Hollebrandse, 2001). Innateness has been regarded as only one of the possible contributing factors.
From a generative perspective, learners map linguistic input onto innate constructs reflecting lexical aspect (e.g., semantic contrasts like telicity) There may be parametric differences between languages with respect to which semantic features are salient in aspectual morphology.
The input immediately reinforces the innate constructs (maybe even activates them), so that learners do not have to spend a lot of time and effort figuring out how lexical aspectual classes are marked in the language they are acquiring. Earlier proposals of this spirit include Bickerton’s (1981; 1984) language bioprogram hypothesis (which proposes the state/process distinction and the punctual/ nonpunctual distinction are pre-programmed) and Slobin’s (1985) basic child grammar hypothesis (which proposes the process/result distinctions are pre-programmed). Li and Shirai’s (2000) work on L1 acquisition of Chinese supports Slobin’s basic child grammar hypothesis rather than Bickerton’s language bioprogram hypothesis, in ascertaining that the telic–atelic distinction is important in children’s early acquisition of tense–aspect markers (p. 119). In their study, Chinese-speaking children in all age groups studied (3 to 6 years) used the imperfective markers almost exclusively with atelic verbs, and the perfective marker predominantly with telic verbs. This pattern becomes even more pronounced in the older children. However, There is no evidence for the bioprogram “state/process” distinction since children understood Chinese perfective marker ‘le’ poorly with both activity and stative verbs and no evidence for the bioprogram “punctual/nonpunctual” since children show no difference in understanding punctual/ non punctual words combined with the progressive marker “zai’ or ‘le’. Studies conducted by Antinucci and Miller (1976) and Olsen and Weinberg (1999) also support Slobin’s (1985) “process/result” distinction. Their studies of children’s production data show that children exhibit divergent patterns from that of adults. Olsen and Weinberg (1999) finds that the English ‘ed’ perfective morpheme is correlated with telic verbs in the child production but not in the adult production. Antinucci and Miller (1976) found that Italian children from 1;6 to 2;6 made the inflectional endings of past participles, which is part of the perfective past tense form passato prossimo, agree with the number and gender of the object of the verb rather than the subject as in the adult language. This suggests that they interpreted the past participle as specifying an attribute of the object rather than an action performed by the subject, and so according to Antinucci and Miller, children at this early stage are not cognitively prepared for past reference unless there is a clear result inherent in the action that serves as a link between the present and the past. In addition, a more recent study of grammatical aspect comprehension by Italian children (van Hout and Hollebrandse, 2001) indicates that the subtle semantics associated with the perfective and imperfective Italian past tenses – the passato prossimo and the imperfetto – have not been acquired at the age of three, although children do use these forms as adults do. Thus, more study regarding the production and comprehension of perfective past tense is warranted to see whether children’s use of perfective past tense is past time reference or perfective marker and where this tendency of present-imperfective and past-perfective correlation exactly originate.
To conclude, the findings of these studies argue against input driven accounts like the connectionist account of the functionalist on the one hand, and argues for the innateness of cognitive saliency of the process-result distinctions as Slobin’s (1985) basic child grammar hypothesis does, on the other hand. But how the existence of the innate process/result distinctions plays out in L2 acquisition remains unclear. To answer this question, we now turn to the next section.
4.Universals in Second Language Acquisition
With regard to second language acquisition of aspect, an important point to note is that the POA framework do not usually take the L2 learners’ native language into account (cf. Slabakova, 2002 for discussion) because it does not predict that typological differences between the L1 and L2 will play a role in the process of aspect acquisition. However, studies conducted by both functionalists and generativists proved otherwise when they take the typological similarity and differences between lexical and grammatical aspect in L1 and L2 into consideration in research.
The present functional studies aim at the revision and expansion of the POA since researchers have demonstrated that the predictive accuracy of the AH varies depending upon different factors and this model is not generalizable to all learners (Bardovi-Harlig, 2000; LiskinGasparro, 2000; López-Ortega, 2000; McManus, 2013; Salaberry, 2003, 2011). Among these factors are L1 influence, L1-L2 structural similarity(Bonilla, 2013; Comajoan, 2005; Comajoan& Pérez-Saldanya, 2005; Izquierdo& Collins, 2008; McManus, 2013; Salaberry, 2011), crosslinguistic difference in lexical aspect (LiskinGasparro, 2000; McManus, 2013; Salaberry, 2002, 2011), and the use of classroom materials versus contact with native speakers(Comajoan,2005; and Salaberry 2002, 2003, 2011).And researchers working within the framework of generative L2 acquisition have paid a great deal of attention to the role of the native language in the acquisition of L2 lexical and grammatical properties and in doing so try to find out to what extent universal grammar guides and restrict this transfer process.
To sum, the typological similarity and differences between lexical and grammatical aspect and their interaction pattern in L1 and L2 serves as the converging point of functionalists and generativists. Thus, different predictions of transfer patterns and interlanguage features can be made according to these two camps’ theoretical assumptions.
In the past decade, research in generative second language (L2) acquisition has focused primarily on issues of syntactic representation, investigating to what extent the principles and parameters of Universal Grammar constrain L2 acquisition. More recently, this focus has broadened to include questions of how knowledge at the interface between syntax and semantics, syntax and the lexicon is acquired, and how knowledge of peripheral mechanisms interacting with, but outside of UG proper, are deployed to instantiate this knowledge (Juffs, 1996; Juffs and Harrington, 1995; Klein and Martohardjono, 1999). Recent research has also begun to explore the semantic aspects of L2 acquisition, investigating the acquisition of lexical semantics and its interaction with argument structure (Juffs, 1996; Hirakawa, 1999, 2001; Inagaki, 1997, Montrul, 1997, 1999; Sorace, 1995, 2000) as well as its interaction with grammatical morphology (Montrul and Slabakova, 2002). Research on the L2 acquisition of aspect falls within this domain, namely, the interaction of lexical semantics and its interaction with grammatical morphology.
Slabakova and Montrul (2002) investigateed the acquisition of the Spanish preterite/imperfective contrast by native speakers of English.L2 learners would have to realize that while English neutralizes the bounded/-unbounded distinction with stative verbs, Spanish does use preterite to mark perfective aspect of stative verbs and denotes complete or bounded events. The L2 learners of Spanish were able to identify the semantic differences in the preterite and imperfect forms, and through positive evidence finound that both grammatical forms are permissible with stative verbs. Thus, they conclude that L2 acquisition is constrained by Universal Grammar and that L2 learners can acquire features of functional categories that are not instantiated in their L1.The Universal Grammar assume a parametric difference between English and Spanish in the feature composition of the functional category Aspectusl Phrase(AspP). In English all verbs (not states) are marked with the feature [+perfective], while in Spanish verbs are not inherently associated with semantic features and the features [+/- perfective] are associated with overt tense morphology instead. Since the two language both have AspP which must be checked by lexical aspect or grammatical aspect, it is proposed that the successful acquisition of aspectual contrasts such as the perfective-imperfective distinction is evidence of semantic feature assignment under the functional category AspP.
Gabriele, Martohardjono and McClure (2005) also evaluate the role of the first language, particularly L1 lexical semantics, in the L2 acquisition of aspect. Based on the fact that the Japanese progressive morpheme ‘te-iru’ allows both progressive and perfective interpretations , depending on the lexical semantics of the verb, whereas English ‘be+ing’ always denotes a progressive interpretation regardless of the verb stem, they proposed the Lexical Semantic Transfer Hypothesis which predicts that in the past progressive better performance will be seen on activity verbs and in the simple past no differences across the two verb classes is predicted. The experiment results indicate that the Lexical Semantic Transfer Model was only partially supported. Contra the predictions of the transfer model, learners did not perform better on activity verbs in the past progressive. Supporting the predictions of the transfer model, learners did perform at equivalent levels with the two verb classes in the simple past. This finding is compatible with the findings of Slabakova and Montrul’s study (2002) in suggesting that transfer does not proceed by verb class and the difficulty of acquisition would lie in the semantics of the functional operators itself, which again tests the universals of mandatory semantic feature assignment under the functional category AspP.
In Haznedar’s (2007) longitudinal study of a Turkish child with at initial stages of L2 English acquisition, the learner initially produced past tense markings on [+punctual] achievement verbs and predicates. Subsequently, her subject began marking [+telic] actions with perfective aspect. The findings of this study, if generalizable, suggest that punctuality and to a lesser degree telicity are more salient to L2 learners than dynamicity at the earliest stages of acquisition of temporal morphology. This contrasts with the POA, which suggests that even early-stage learners make use of all three semantic features to determine lexical aspect.
Slabakova (2015) reports on an experimental study that English learners utilize the universal deictic pattern in their comprehension of the temporal information of the tenseless language Mandarin, which postulates that unbounded situations are located in the present and bounded events are located in the past (Smith &Erbaugh, 2005).That is, the lack of explicit tense marking in Chinese doesn’t pose extra burden to English learners and its abundances of aspect markers are suffice to allow acquisition. This study further attests the role of universals in L2 aspect acquisition.
To sum, the aforementioned studies serve as testimony for the existence of universals grammar as facilitator in L2 aspect acquisition, especially via the mandatory semantic feature assignment under the functional category AspP. However, it remains unclear what the parameters are regarding the crosslinguistic differences in how particular lexical classes interact with grammatical morphology and where the features such as [+/- perfective] are checked, whether in lexical aspect or grammatical aspect or both.
5.Summary of findings and directions for future research
In summary, the current body of research finds that universal grammar does play a role in L1 and L2 aspect acquisition. But it remains unclear how the supposed innateness of process and result distinction manifested in L1 acquisition plays out in L2 acquisition. Moreover, although current study favors the idea that crosslinguistic differences in how particular lexical classes interact with grammatical morphology lies in the different semantic features of grammatical morphology rather than aspectual information encoded in verbs, further study is needed to gain more evidence of this proposal.
In addition, in languages where deictic past and aspectual meanings in their verbal morphology is conflated, more study regarding the production and comprehension of perfective past tense is warranted to see whether children’s use of perfective past tense is past time reference or perfective marker and whether the interlanguages with regard to these languages are tense or aspect oriented.
References：
Andersen, R. W. (1991). Developmental sequences: The emergence of aspect marking in second language acquisition. In Huebner, T., & Ferguson, C. A. (eds.), Crosscurrents in SLA and Linguistic Theories (pp. 305–324). Amsterdam & Philadelphia: John Benjamins.
Andersen, Roger W. (1993). Four operating principles and input distribution as explanations for underdeveloped and mature morphological systems. In Hyltenstam, K., &Viberg, A. (eds.), Progression and Regression in Language (pp. 309–339). Cambridge: Cambridge University Press.
Andersen, R., W. & Shirai, Y. (1994). Discourse motivations for some cognitive acquisition principles. Studies in Second Language Acquisition 16(2). 133–156.
Andersen, R., W. & Shirai, Y. (1996). The primacy of aspect in first and second language acquisition: The pidgin-creole connection. In Ritchie, W. C., & Bhatia T. K. (eds.), Handbook of Second Language Acquisition (pp. 527–570). San Diego, CA: Academic Press.
Bardovi-Harlig, K. (1998). Narrative structure and lexical aspect: Conspiring factors n second language acquisition of tense- aspect morphology. Studies in Second Language Acquisition, 20, 471-508.
Bardovi-Harlig, K. (1999). From morpheme studies to temporal semantics. Studies in Second Language Acquisition, 21, 341–82.
Bardovi-Harlig, K. (2000). Tense and aspect in second language acquisition: Form, meaning, and use. Oxford: Blackwell.
Chen, J., & Shirai, Y. (2015). The acquisition of relative clauses in spontaneous child speech in Mandarin Chinese. Journal of child language, 42(2), 394-422.
Comrie, B. (1976). Aspect. New York: Cambridge University Press.
Jin, L., & Hendriks, H. (2005). The development of aspect marking in L1 and L2 Chinese. Working Papers in English and Applied Linguistics, 9, 69-99.
Kumpf, L. (1984). Temporal systems and universality in interlanguage: A case study. In Eckman, F., Bell, L., & Nelson, D. (Eds.), Universals of Second Language Acquisition (pp. 132-143). Rowley, MA: Newburry House.
Li, P. and Shirai, Y. (2000). The acquisition of lexical and grammatical aspect. Berlin: Mouton de Gruyter
Li, P., & Bowerman, M. (1998). The acquisition of lexical and grammatical aspect in Chinese. Language, 18(54), 311-350.
Luk, Z. P. S., & Shirai, Y. (2018). The development of aspectual marking in Cantonese-English bilingual children. International Review of Applied Linguistics in Language Teaching, 56(2), 137-179.
Nishi, Y., & Shirai, Y. (2016). The role of linguistic explanation in the acquisition of Japanese imperfective-teiru. In Theory, research and pedagogy in learning and teaching Japanese grammar (pp. 127-155). Palgrave Macmillan, London.
Olsen, M.B. (1997). A semantic and pragmatic model of lexical and grammatical aspect. New York: Garland.
Olsen, M.B. and Weinberg, A. (1999). Innateness and the acquisition of grammatical aspect via lexical aspect. In Greenhill, A., Littlefield, H. and Tano, C., editors, Proceedings of the 23rd annual Boston University conference on language development. Somerville, MA: Cascadilla Press, 529–40.
Pi?ango, M., Zurif, E. and Jackendoff, R. (1999). Real-time processing implications of enriched composition at the syntax-semantics interface. Journal of Psycholinguistic Research 28, 395–414.
Qian, C. (2015). A Cross-Sectional Study on the Roles of Lexical Aspect and L1 Transfer in Tense-Aspect Acquisition. English Language Teaching, 8(7), 114-132.
Robison, R. E. (1995). The aspect hypothesis revisited: A cross-sectional study of tense and aspect marking in interlanguage. Applied Linguistics, 16, 344-370.
Rong, H. (2006). Explaining the Aspect Hypothesis: Input and Semantic Prototype. Beijing: Foreign Language Education.
Slabakova, R. (2015). Acquiring Temporal Meanings Without Tense Morphology: The Case of L2 Mandarin Chinese. The Modern Language Journal, 99(2), 283-307
Ryu, J. Y., Horie, K., & Shirai, Y. (2015). Acquisition of the Korean Imperfective Aspect Markers–ko iss–and–a iss–by Japanese Learners: A Multiple‐Factor Account. Language Learning, 65(4), 791-823.
Salaberry, M.R. (2000). The development of past tense morphology in L2 Spanish. Philadelphia: John Benjamins.
Shirai, Y. (2014). The semantic basis of grammatical development: Its implications for modularity, innateness, and the theory of grammar. In Usage-based Approaches to Japanese Grammar (pp. 153-170). John Benjamins.
Shirai, Y., & Andersen, R. W. (1995). The acquisition of tense-aspect morphology: A prototype account. Language, 71(4). 743–761.
Shirai, Y., & Juffs, A. (2017). Introduction: Convergence and divergence in functional and formal approaches to SLA. Second Language Research, 33(1), 3-12.
Slabakova, R. (2002). Recent research on the acquisition of aspect: an embarrassment of riches? Second Language Research, 18(2), 172-188.
Slabakova, R. 2001. Telicity in the second language. Amsterdam: John Benjamins.
Slabakova, R. & Montrul, S. ( 2002). On viewpoint aspect interpretation and its L2 acquisition: a UG perspective. In Shirai, Y. and Salaberry, R., editors, Tense–aspect morphology in L2 acquisition. Amsterdam: John Benjamins.
Smith, C. (1997). The Parameter of Aspect (2nd edition). Dordrecht: Kluwer Academic Publishers.
Soh, H. L., &Kuo, J. Y. C. (2005). Perfective aspect and accomplishment situations in Mandarin Chinese. In Perspectives on aspect (pp. 199-216). Springer, Dordrecht.
Sun, H. (2014). Temporal construals of bare predicates in Mandarin Chinese. LOT: Utrecht.
Thane, P. D. (2018). The present state of the aspect hypothesis: A critical perspective. Eurasian Journal of Applied Linguistics, 4(2), 261-273.
Vendler, Z. (1967). Verbs and times. In Z. Vendler (Ed.), Linguistics and Philosophy (pp. 97–121). Ithaca, NY: Cornell University Press. (Reprinted from Philosophical Review, 1957, 66, 143–160).
Laura,W. (2001)Aspectual influences on early tense comprehension. Journal of Child Language, 28(3), 661-681.
Wulff, S., Ellis, N. C., R?mer, U., Bardovi–harlig, K., & Leblanc, C. J. (2009). The acquisition of tense–aspect: Converging evidence from corpora and telicity ratings. The Modern Language Journal, 93(3), 354-369.