Cory Bill | Macquarie University (original) (raw)

Papers by Cory Bill

2 3 Presuppositions (Ps) and Scalar implicatures (SIs) are traditionally considered to be distinc... more 2 3 Presuppositions (Ps) and Scalar implicatures (SIs) are traditionally considered to be distinct 4 phenomena. But recently, various authors have argued that (at least certain) Ps should be 5 analyzed as SIs to capture differences between presupposition triggers. This scalar implicature 6 approach to presuppositions predicts that, all else being equal, the two types of inferences 7 will display uniform behavior. Some initial processing studies tested this prediction, but yielded 8 conflicting results. While some found no delay in Reaction Times (RTs) associated with Ps 9 under negation, which is in contrast with previous findings for SIs and thus challenges the SI 10 approach to Ps, others compared SIs and Ps under negation directly and found uniform RT 11 patterns for them. We review the conflicting results and argue that the difference in outcomes 12 is attributable to a difference in the type of response being measured: the RTs associated with 13 acceptance and rejection judgments seem to pattern in exactly opposite ways. Next, we report 14 on two experiments that systematically look at both acceptance and rejection responses for 15 SIs and Ps respectively. The results suggest that, once the acceptance vs. rejection variable 16 is factored in, RT patterns for Ps and SIs are indeed uniform, and only involve delays for the 17 respective inferences when looking at rejection judgments. However, a contrast between Ps and 18 SIs re-emerges in the results of the third experiment, in which auditory stimuli are used to assess 19 the impact of prosody on deriving the respective inferences. This poses a first challenge to the 20 scalar implicature approach to presuppositions. A third set of experiments investigates a further 21 prediction by this approach by looking at a relevant presupposition trigger in affirmative contexts. 22 According to the SI approach to Ps, the presuppositional inference is introduced as a simple 23 entailment in affirmative contexts. This predicts that language users' behaviour towards other 24 entailments and these presuppositional inferences will be uniform. We compare rejections of 25 affirmative sentences based on either their presuppositional inference or their entailed content, 26 and found longer RTs for the former case. A final variation investigated whether there might 27 be a parallel difference in the RTs of two entailments associated with always, and found no 28 such difference. The results produced by the second two sets of experiments pose a substantial 29 challenge to approaches assimilating Ps and SIs, while being entirely in line with the traditional 30 perspective of seeing the two phenomena as distinct in nature. 31

Previous developmental studies have revealed variation in children's ability to compute scalar in... more Previous developmental studies have revealed variation in children's ability to compute scalar inferences. While children have been shown to struggle with standard scalar inferences (e.g., from scalar quantifiers like some) (Noveck

Plural morphology in English is associated with a multiplicity inference, e.g., "Emily fed giraff... more Plural morphology in English is associated with a multiplicity inference, e.g., "Emily fed giraffes" is typically interpreted to mean that Emily fed more than one giraffe. It has long been observed that this inference disappears in downward-entailing linguistic environments, such as in the scope of negation. There are two main approaches to this puzzle: the first proposes that the plural is ambiguous, and invokes a preference for stronger meanings (Farkas & de Swart 2010), while the second derives multiplicity inferences as implicatures (Spector 2007; Zweig 2009; Ivlieva 2013; Mayr 2015). In this paper, we report on two experiments comparing how adults and preschool-aged children interpret plural morphology. The first experiment reveals that both adults and 4–5-year-old English-speaking children compute more multiplicity inferences in upward-entailing environments than in downward-entailing environments, and moreover that children compute fewer multiplicity inferences than adults. The second experiment tested a new group of participants on both multiplicity inferences and the upper-bound scalar implicature of the quantifier "some". Again, both groups computed more multiplicity inferences in upward-entailing than in downward-entailing environments. More importantly, children computed fewer multiplicity inferences and scalar implicatures than adults, and their performance on the two kinds of inferences was significantly correlated. We discuss how the findings of the two experiments support a scalar implicature approach to multiplicity inferences, while posing a challenge for the ambiguity approach.

The nature of presupposition projection from the scope of quantificational expressions, particula... more The nature of presupposition projection from the scope of quantificational expressions, particularly negative ones, continues to be both theoretically and empirically controversial. We experimentally investigate the interpretation of sentences like None of the bears won the race, which could in principle be associated with three projection options, giving rise to: an existential reading, on which at least one of the bears participated and none won; a universal reading, on which all of the bears participated and none won; and a presuppositionless reading on which none of the bears both participated and won. We used a covered box task to investigate English-speaking adults' and children's interpretations of such sentences. The results from the adult participants provide empirical evidence for all three readings; the results from the child participants, however, reveal no evidence for the existential reading. We discuss our results in light of competing theoretical accounts, focusing on options for reconciling the theories with the experimental data from both adults and children.

Trends in Acquisition (TiLAR): Semantics in Acquisition

In natural language, we encounter various sentence types that, under certain circumstances, are e... more In natural language, we encounter various sentence types that, under certain circumstances, are evaluated as neither true nor false. For instance, it is intuitively difficult to assess the truth value of a sentence whose presupposition is not satisfied in the context. A common theoretical approach is to characterize the status of such sentences with a third value of one kind or another. In this chapter, we consider children’s acquisition of four linguistic phenomena that can give rise to 'gappy' judgments that correspond neither to True nor False: scalar implicature, presupposition, homogeneity, and vagueness. We discuss how young children's interpretations of such sentences can provide insight into how these phenomena should be treated within semantic theories.

There exist various sentence types in natural language that, under certain circumstances, are eva... more There exist various sentence types in natural language that, under certain circumstances, are evaluated as neither true nor false. For instance, in a context in which the presupposition of a sentence is not satisfied, it is intuitively rather difficult to assess what the truth value of the sentence should be. A common theoretical approach is to characterize the status of such a sentence with a third value of one kind or another. In this paper, we consider children’s ac- quisition of four linguistic phenomena that can give rise to ‘gappy’ judgments that correspond neither to True nor False: scalar implicature, presupposition, homogeneity, and vagueness. We discuss how young children’s interpretations of such sentences can provide insight into how these phenomena should be treated within semantic theories.

The purpose of this paper is to evaluate different explanations of a particular pronoun error whi... more The purpose of this paper is to evaluate different explanations of a particular pronoun error which children sometimes produce during language acquisition. This error involves children producing non-nominative pronouns in the subject position of some sentences (e.g. Her is kicking the ball). There has been an explanation of this phenomenon (known as the Agreement/Tense Omission Model, or ATOM) put forward by Schütze & Wexler (1996), based on the Universal Grammar theory. This account has been challenged in a number of papers (Pine, Rowland, Lieven, & Theakston, 2005; Ambridge & Pine, 2006), which found data that was not consistent with the predictions of the ATOM. Pine et al. (2005) went on to suggest that this pronoun error phenomenon could be better explained through a Constructivist or input-based theory. This paper endeavours to evaluate this claim, by investigating whether the language input provided to the children in Pine et al.'s (2005) study is consistent with a number ...

Since Bott and Noveck (2004), there has been an ongoing discussion about whether scalar implicatu... more Since Bott and Noveck (2004), there has been an ongoing discussion about whether scalar implicatures are delayed in online processing relative to literal meaning. Bott and Noveck (2004) provided Reaction Time evidence for such a delay, replicated in a number of later variations of their study (e.g., Bott et al., 2012). Breheny et al. (2006) found corresponding delays in self- paced reading. More recently, the issue has been investigated using the visual world paradigm, where results have been more mixed. Huang and Snedeker (2009, and subsequent work) have found delays for eye movements based on the ‘not all’ implicature of ‘some.’ But various others, (e.g., Grodner et al. (2010), Breheny et al. (2013), and Degen and Tanenhaus (2011)), report results which they argue show that implicatures are available immediately. Schwarz et al. (2015) added another angle to this picture, by using a sentence picture matching task using a Covered Picture (or Covered Box; henceforth CB; Huang et al., 2013), that allowed RT comparisons both within ac- ceptance (target) and rejection (CB) responses. While replicating the delay for implicature-based rejection responses, they find the reverse pattern for acceptance responses, with faster RTs for implicature-compatible conditions. They propose that delays associated with literal acceptances and implicature-based rejections result from a conflict between the two possible interpretations, rather than reflecting a cost of implicature-calculation. The present experiment extends this ap- proach beyond RTs by combining Visual World eye-tracking with the CB paradigm. The results a) are consistent with the notion that both literal and implicature interpretations are available in par- allel; b) show that literal acceptances are nonetheless only provided reluctantly, presumably due to a preference for implicature meanings, in line with Schwarz et al’s proposal; and c) suggest that for both literal acceptances and implicature-based rejections, there is a competition effect between the two interpretations. In addition, the RT data display an implicature-based block-priming effect, suggesting that the resolution of this conflict can be sped up through repeated task-exposure.

Scalar implicatures (SIs) and Presuppositions (Ps) are traditionally con- sidered to be distinct ... more Scalar implicatures (SIs) and Presuppositions (Ps) are traditionally con- sidered to be distinct phenomena. But recently, various authors have argued that (at least certain) Ps should be analyzed as SIs to capture differences between pre- supposition triggers (e.g., Abusch 2002, 2010, Simons 2001, Chemla 2009, 2010, Romoli 2012, 2014). This scalar implicature approach to presuppositions predicts that, all else being equal, the two types of inferences will display uniform behav- ior. Some initial processing studies tested this prediction, but yielded conflicting results. Chemla & Bott 2013 found no delay in Reaction Times (RTs) associated with Ps under negation, which is in contrast with previous findings for SIs and thus challenges the SI approach to Ps. However Romoli & Schwarz 2015 compared SIs and Ps under negation directly and found uniform RT patterns for them. We review the conflicting results and argue that the difference in outcomes is attributable to a difference in the types of responses whose RTs are measured: acceptance and rejection judgments seem to pattern in exactly opposite ways. Next, we report two experiments that systematically look at both acceptance and rejection responses for SIs and Ps respectively. The results of these experiments suggest that, once the acceptance vs. rejection variable is factored in, RT patterns for Ps and SIs are indeed uniform, and only involve delays for the respective inferences when looking at rejection judgments. This outcome pattern undermines the general case for delays in online implicature computation based on reaction times, and we outline an alter- native perspective on the data. And, importantly, it is entirely in line with the scalar implicature approach to presuppositions. Another aspect of this approach, however, is that in affirmative contexts, the presuppositional inference is a simple entailment. This, in turn, predicts that entailments and presuppositions should behave uniformly in such contexts. We report on a second series of experiments testing this prediction, following the approach in Kim 2007 and Schwarz 2015a. We looked at rejections of sentences based on either their presupposition or their assertion, and found longer RTs for the former case. This, together with other differences between Ps and SIs found in studies on language acquisition and language disorders (Bill et al. 2014, Kennedy et al. 2015), does pose a substantial challenge to scalar implicature ap- proaches to presupposition, while it is entirely in line with the traditional perspective of seeing the two phenomena as distinct in nature.

Plural morphology in English is associated with a ‘more than one’ inference, which we refer to a... more Plural morphology in English is associated with a ‘more than one’ inference, which we refer to as a plurality inference, e.g., Emily fed giraffes 􏰀 Emily fed more than one giraffe. As with scalar implicatures, the plurality inference disappears in downward- entailing environments, such as when a plural appears in the scope of negation. This has led to the proposal that it should be derived as a kind of scalar implicature (Spec- tor 2007; Zweig 2009; Ivlieva 2013; Magri 2014). In this paper, we investigated this proposal in the domain of child language. There were two experiments. The first was conducted with adults and with 4–5-year-old children. The main finding was that both groups computed more plurality inferences in upward-entailing linguistic environments than in downward-entailing environments. Moreover, children computed fewer plural- ity inferences overall than adults did. This finding is consistent with previous research demonstrating children’s relative insensitivity to scalar implicatures. Taken together, the findings of the first experiment were consistent with a scalar implicature approach to plurality inferences. The second experiment tested different groups of participants on both plurality inferences and on classical scalar implicatures. Again, both the child and adult participants computed more plurality inferences in upward-entailing than in downward-entailing environments and, again, children computed fewer plurality infer- ences than adults did. More importantly, children’s performance on plurality inferences was highly correlated with their performance on classical scalar implicatures. We dis- cuss the implications of the findings for theories of plurality inferences, and for the acquisition of plurality more generally.

The present study focuses on the interpretation of plurality inferences by individuals with Broca... more The present study focuses on the interpretation of plurality inferences by individuals with Broca's aphasia. The main finding is they successfully computed plurality inferences (PIs), and they did so significantly more often in upward entailing (UE) than downward entailing (DE) contexts, in parallel with the typical controls. This competence in computing PIs is in keeping with the correct use of plural morphology observed in the production of people with Broca's aphasia, but it is surprising given their clear impairments in understanding sentences with complex syntactic structures.

This paper reports on an experimental investigation of presuppositions and im- plicatures in lang... more This paper reports on an experimental investigation of presuppositions and im- plicatures in language acquisition. The comparison bears on an important the- oretical issue, as recent theoretical proposals (Chemla 2009, Romoli 2012, 2014 among others) posit the same mechanisms for both types of inferences, in con- trast with the traditional view. We used a modified ‘Covered Box’ task (Huang, Spelke and Snedeker 2013) to compare direct scalar implicatures (some infers not all) and indirect scalar implicatures (not all infers some) to presuppositions (didn’t win infers participated), in two groups of children (4–5 year olds & 7 year olds) and one group of adults. The main finding is that children’s and adults’ behavior differs across inference types. This result challenges the recent uniform accounts but is in line with the traditional distinction between these inferences. Finally, we also found an unexpected distinction between the two types of scalar implicatures, which is not predicted by any theory we are aware of. We discuss the results as they relate to recent sentence processing results, as well as to the ongoing theoretical discussion around the nature of these inferences, and how they are generated.

in proceedings of SuB 19

The processing of scalar implicatures has been extensively studied in recent years (Bott and Nove... more The processing of scalar implicatures has been extensively studied in recent years (Bott and Noveck 2004; Bott et al. 2012; Breheny et al. 2006; Huang and Snedeker 2009, Chemla and Bott 2013 among others). Following the work of Bott and Noveck (2004), there has been an ongoing discussion about whether scalar implicatures are delayed in online processing. A second question, discussed more recently, is whether so-called ‘indirect’ scalar implicatures (e.g., John didn’t always go to the movies 􏰀 John sometimes went) exhibit the same processing profile as the standard direct ones (e.g., John sometimes went to the movies 􏰀 John didn’t always go). Cremers and Chemla (2014) argue that they do, whereas Romoli and Schwarz (2015) find the reverse pattern for indirect scalar implicatures, with slower Reaction Times associated with literal responses. In this paper, we report an experiment investigating two key questions arising from this debate: first, do Reaction Times yield uniform evidence for delayed availability of implicatures? and second, do direct and indirect scalar implicatures display comparable or distinct processing properties? Using the covered picture version of a picture matching task (Huang et al. 2013) with reaction time measures, we look both at cases where the overt picture is rejected and where it is accepted. Our results provide a negative answer to the first question: while delays for implicature responses arise for rejections (Covered picture choices), the pattern flips for acceptances (Target choices). The cross-over interaction in the results is inconsistent with attributing delayed implicature-rejections to delays in the availability of the implicature, since Target choices compatible with the implicature are faster than ones that are only compatible with a literal interpretation. This yields a very different perspective on reaction time results for implicatures. As for the second question, our results show that once acceptance vs. rejection is factored in, the general reaction time pattern is the same for indirect and direct scalar implicatures and compatible with the different results found in the literature.

This paper reports an experimental investigation of presuppositions and scalar implicatures in la... more This paper reports an experimental investigation of presuppositions and scalar implicatures in language acquisition. Recent proposals (Chemla 2009; Romoli 2012, 2014) posit the same mechanisms for generating both types of inferences, in contrast to the traditional view. We used a Covered Box picture selection task to compare the interpretations assigned by two groups of children (4/5 and 7 year olds) and by adults, in response to sentences with presuppositions and ones with either ‘direct’ or ‘indirect’ scalar implicatures. The main finding was that the behavior of children and adults differed across inference types. This asymmetry is consistent with the traditional perspective, but poses a challenge for the more recent uniform accounts. We discuss how the latter could be amended to account for these findings, and also relate the findings to previous results on presupposition processing. Finally, we discuss an unexpected difference found between direct and indirect scalar implicatures.

in proceedings of SALT 24

This paper provides novel experimental evidence for a scalar implicature approach to the pluralit... more This paper provides novel experimental evidence for a scalar implicature approach to the plurality inferences that are associated with English plural morphol- ogy (Emily fed giraffes 􏰀 Emily fed more than one giraffe). Using a Truth Value Judgment Task, we show that both adults and 4–5-year-old children compute more plurality inferences in upward-entailing than downward-entailing environments, but children compute fewer plurality inferences overall than adults do. These findings are consistent with previous research demonstrating children’s relative insensitivity to scalar implicatures. We discuss the implications of these findings for theories of plurality inferences, and for the acquisition of scalar inferences more generally.

Conference Presentations by Cory Bill

This paper reports on an experimental investigation of presuppositions and implicatures in langua... more This paper reports on an experimental investigation of presuppositions and implicatures in language acquisition. The comparison bears on an important the- oretical issue, as recent theoretical proposals (Chemla 2009, Romoli 2012, 2014 among others) posit the same mechanisms for both types of inferences, in con- trast with the traditional view. We used a modified ‘Covered Box’ task (Huang, Spelke and Snedeker 2013) to compare direct scalar implicatures (some infers not all) and indirect scalar implicatures (not all infers some) to presuppositions (didn’t win infers participated), in two groups of children (4–5 year olds & 7 year olds) and one group of adults. The main finding is that children’s and adults’ behavior differs across inference types. This result challenges the recent uniform accounts but is in line with the traditional distinction between these inferences. Finally, we also found an unexpected distinction between the two types of scalar implicatures, which is not predicted by any theory we are aware of. We discuss the results as they relate to recent sentence processing results, as well as to the ongoing theoretical discussion around the nature of these inferences, and how they are generated.

Talks by Cory Bill

Introduction The nature of presupposition projection from the scope of quantificational sentences... more Introduction The nature of presupposition projection from the scope of quantificational sentences , particularly negative quantificational sentences such as (1), continues to be both theoretically and empirically controversial. Standard candidates for the readings of (1) are the following:

Trends in Acquisition (TiLAR): Semantics in Acquisition

in proceedings of SuB 19

in proceedings of SALT 24

Bill et al. (2014) report that 4-year-olds have knowledge of the global presupposition of win (1a... more Bill et al. (2014) report that 4-year-olds have knowledge of the global presupposition of win (1a), but struggle with local accommodation (1b).
(1) The bear didn’t win the race.
a. Presupposition: The bear participated in the race.
b. Local accommodation: The bear didn’t both participate and win the race.
In this study, we embed the presuppositional trigger under the negative quantifier none (2). The nature of such presupposition projection has been controversial, with the readings in (2) being variously endorsed by different authors (Heim 1983, Beaver 1994, Schlenker 2008). We investigate the availability of these readings, focusing on whether children are able to access the same set of interpretations as adults.
(2) None of the bears won the race.
a. Existential presupposition: At least one bear participated in the race.
b. Universal presupposition: All of the bears participated in the race.
c. Local accommodation: None of the bears both participated and won the race.

Previous research has found that children typically do not compute the scalar impli- catures (SIs... more Previous research has found that children typically do not compute the scalar impli- catures (SIs) associated with sentences containing a single scalar term, such as (1a). This study investigated children’s comprehension of sentences containing multiple scalar terms, such as (2a). The main finding was that with such sentences, children computed SIs on a par with adults. We suggest that children succeeded in computing SIs for two reasons: (i) the scalar alternative every appeared in the sentence; (ii) the visual context explicitly depicted the truth conditions associated with the alternative scalar terms some and every.

Recently a scalar implicature approach has been applied to plurality inferences. To evaluate this... more Recently a scalar implicature approach has been applied to plurality inferences. To evaluate this unified approach, the present study compared children’s and adults’ comprehension of sentences containing plural morphology and sentences containing the scalar term some. There were two main findings. First, both children and adults computed more plurality inferences (PIs) in upward-entailing than in downward- entailing linguistic environments. Second, children computed much fewer PIs and scalar implicatures from sentences with some (SIs) than adults did. The findings are consistent with previous research demonstrating children’s relative insensitivity to scalar inferences, and provide support for an SI approach to PIs.

Since Bott & Noveck 2004 (B&N), there has been an ongoing discussion about whether scalar implica... more Since Bott & Noveck 2004 (B&N), there has been an ongoing discussion about whether scalar implicatures are delayed in online processing relative to literal meaning. B&N provided Reac- tion Time (RT) evidence for such a delay, replicated in a number of later variations of their study (Bott et al. 2012, Bott and Chemla 2013 a.o.). Breheny et al. (2006) found corresponding de- lays in self-paced reading. More recently, the issue has been investigated using the visual world paradigm (VW), where results have been more mixed. Huang & Snedeker (2009, and subsequent work) (H&S) have found delays for eye movements based on the ‘not all’ implicature of ‘some.’ But various others, e.g., Grodner et al. (2010), Breheny et al. (2013), and Degen & Tanenhaus (2011), report results which they argue show that implicatures are available immediately. Schwarz et al. (2014) added another angle to this picture, by using a sentence picture matching task using a Covered Picture (or Covered Box; henceforth CB; Huang et al. 2013), that allowed RT compar- isons both within acceptance (target) and rejection (CB) responses. While replicating the delay for implicature-based rejection responses, they find the reverse pattern for acceptance responses, with faster RTs for implicature-compatible conditions. They propose that delays associated with literal acceptances and implicature-based rejections result from a conflict between the two possible inter- pretations, rather than reflecting costs of implicature-calculation. The present experiment extends this approach beyond RTs by combining Visual World eye-tracking with the CB paradigm. The results a) are consistent with the notion that both literal and implicature interpretations are rapidly available; b) show that literal acceptances are nonetheless only provided reluctantly, presumably due to a preference for implicature meanings, in line with Schwarz et al.’s proposal; and c) suggest that for both literal acceptances and implicature-based rejections, there is a competition effect be- tween the two interpretations. In addition, the RT data display an implicature-based block-priming effect, suggesting that the resolution of this conflict can be sped up through repeated task-exposure.

This study compares individuals with Broca’s aphasia (IWBA), 7-yr-old children, and a control gro... more This study compares individuals with Broca’s aphasia (IWBA), 7-yr-old children, and a control group of adults in an experiment involving scalar implicatures and presuppositions in negative sentences. The main findings were that the IWBA were adult-like in computing scalar implicatures (SIs), whereas they were the equals of the group of 7-yr-olds in computing presuppositions (Ps). The P-result provides novel evidence for processing limitations outside the domain of syntax in IWBA. The SI-result suggests either (a) SIs do not incur a processing cost, contrary to much of the literature, or (b) the processing cost associated with computing SIs is significantly less than that associated with Ps, such that SIs are unproblematic for IWBA. From a theoretical perspective, the observed differences between SIs and Ps are consistent with the traditional view, which treats them as distinct linguistic phenomena, but are at least prima facie at odds with recent proposals that have attempted to provide a unified analysis of both phenomena (e.g., Romoli 2014

Selected References • Sauerland et al 2005. The plural is semantically unmarked. • Spector 2007. ... more Selected References • Sauerland et al 2005. The plural is semantically unmarked. • Spector 2007. Aspects of the pragmatics of plural morphology. • Pearson et al 2011. Even more evidence for the emptiness of plurality.