1. Introduction

Gil, Marsden and Whong (henceforth GMW, where reference to a specific paper is not required) have explored the acquisition of existential quantifiers in a range of second languages by speakers of different first languages (; ; ). The acquisition of paradigms such as English any, or Korean wh-existential nwukwu (‘who/anyone/someone’) is of theoretical interest for various reasons. First, their grammatical licensing and semantic interpretation are subject to subtle constraints. Patterns of acquisition thus furnish insight into how learners may overcome learnability issues at the syntax-semantics interface. Second, and linked to this, is the question of L1 influence in the acquisition of L2 grammars, given only partial overlap in the distribution and interpretation of related paradigms in different L1-L2 pairings. L2 learners’ acquisition therefore furnishes insights into the nature of transfer and grammar restructuring in response to input patterns (see , based on ). And third, some properties of these items are taught in foreign language instruction, so that one may be able identify effects of explicit instruction in connection with, or in competition with, properties of acquisition.

In the present study, a paced acceptability judgment task from Marsden et al. () was partly replicated. The replication complements the existing research by adding evidence from learners with L1 German. The syntactic and semantic properties of the relevant German features differ markedly from the L1s previously tested, resulting in further evidence relevant to potential L1 influence. The sentence types tested in the study reflected properties of the English existential quantifier any as a negative polarity item (NPI). These properties guided the research questions and hypotheses. We ask whether observable and/or learned grammar properties shape L2 knowledge, and in how far this L2 knowledge was potentially influenced by L1 transfer.

We proceed as follows. First, we will give an account of any as a negative polarity item and then summarise studies on the L2 acquisition of NPI any. Subsequently, we outline properties of related German existential quantifier paradigms. This serves as the basis for presentation of the experiment. Discussion of the results and suggestions of potential avenues for further research close the paper.

1.1 The Distribution of NPI any in English

In English, any as an NPI requires a negative environment. Typically, this means occurring in the scope of sentential negation, as in (1). The occurrence of any outside the scope of negation is ungrammatical, as in (2).

(1)  I did not read any book.
(2)*Anyone did not read the book.

Beyond this basic case, there are various subtleties regarding the precise characterisation of NPI licensing, and how semantic properties interact with the lexical requirements of NPIs, which are more or less stringent in the type of negative context required (see ). To exemplify, any is licensed in questions, see (3). While questions do not involve overt negation, they can be analysed as related to nonveridicality. That is, involving the truth value of the content of the clause, or assertion of whether the event or state of affairs under consideration actually happened/will happen (see ; ).

(3)Has anyone already read the book? (cf. *Anyone read the book.)

Any also occurs without overt negation in contexts where a nonveridical or negative inference can be computed (see ). It is therefore licensed in clauses embedded under a semantically negative verb or within the scope of semantically negative adverbs, as in (4) and (5) respectively, from Gil et al. ().

(4)a.  John regrets that he ate anything at the party.
 b.*John thinks that he ate anything at the party.

(5)a.  John hardly ate anything at the party.
 b.*John probably ate anything at the party.

In L2 English, acquisition involves two related issues. Firstly, the lexical representation of any must specify it as a negative polarity item, so that its distribution is restricted. In syntactic terms, one could propose that this means any bears an NPI feature (), or an uninterpretable nonveridical feature (), which is checked in specific syntactic configurations. The second issue involves establishing what these configurations are, i.e. that an overt negative operator such as not, an interrogative operator, or an implicit negator introduced by the semantics of certain verbs and adverbs can create the requisite grammatical environments.

1.2 The Acquisition of any in L2 English

The use of any is often addressed in instruction in English as a foreign language (EFL, e.g., ). However, the information typically provided in instruction underdetermines the range of properties outlined above. This leads GMW to ask how L2 knowledge of a linguistic phenomenon develops when certain properties are taught, others are not taught but may be observable in incidental input, and still others are neither taught nor observable in input.

Based on studies of online EFL teaching materials () and international EFL textbooks (), it is shown that the typical pedagogical approach to any involves contrasting its use with some and characterising the choice as determined by clause type. Any occurs in questions and negative declaratives, while some occurs in affirmative declaratives. Additional information may be provided as ‘exceptions’ to this basic rule, i.e., that certain adverbs collocate with any. Learners who have received such instruction would be able to correctly accept any with sentential negation and in questions (assuming such knowledge is accessible during performance). However, an overgeneralisation of this rule of thumb would complicate knowledge of more subtle properties. The absence of overt negation, as in (4a) and (5a) would lead learners to reject these sentences if they are relying on the ‘textbook-rule’ of using any with overt negation. Similarly, the presence of negation in a sentence like (2) could conceivably mislead learners into accepting this pattern due to the occurrence of a negator, even though the scope properties in this configuration rule it out.

Using paced acceptability judgement tasks, GMW investigated the acquisition of NPI properties of any by L1 Arabic and L1 Chinese-speaking learners. The aims of these studies were to explore whether learners’ judgements indicate an ability to reject the less observable properties of any’s distribution dependent on scope and licensing from implicit negation, and whether learners might potentially overgeneralise textbook rules. In Marsden et al. (), the judgement task elicited acceptability ratings for 8 sentence types, reflecting the NPI properties and pedagogical rules discussed above, with both grammatical and ungrammatical stimuli. In their study, 4 tokens of each type resulted in 32 experimental items, plus 32 fillers. Table 1 summarises result for the eight sentence types.

Table 1

Sentence types for acceptability judgement tasks, adapted from Marsden et al. (), and mean accuracy of acceptability judgement tasks by sentence type, adapted from Gil et al. ().


L1 ARABIC (N = 25)L1 CHINESE (N = 22)L1 ENGLISH (N = 15)

questioninterrogative operator, non-veridicalgrammaticalDo you have any homework today?3.843.863.93

affirmative declarativeungrammatical*I’ve heard any news about the campaign3.082.863.73

negative declarativenegative operatorgrammaticalThe teacher did not set any homework3.683.914.00

declarative outside scopeungrammatical*Anyone did not follow the instructions2.322.683.87

negative verbimplicit negationgrammaticalI’m sorry I said anything about your driving test2.882.233.73

non-factive verbungrammatical*I guess that you know anything about my visit2.122.093.60

negative adverbimplicit negationgrammaticalJames hardly ate anything at the party2.923.363.93

possibility adverbungrammatical*Lucy probably bought anything last week2.522.413.73

All groups had lower accuracy on ungrammatical sentence types, although the results of t-tests on the grammatical/ungrammatical pairs were not significant for the L1 English group, indicating that they were equally able to accept grammatical clauses and reject ungrammatical ones. They also generally had higher accuracy across the board compared to learners. Overall, both learner groups performed similarly. Accuracy was highest on those contexts which tested the typically taught pedagogical rules: the difference between questions and declaratives and the fact that any occurs with the sentential negator not. Judgements are less consistent on those contexts which are not covered in pedagogical materials. However, this does not demonstrate that the pedagogical rule is overgeneralised. Such overgeneralisation would be evidenced by rejection of sentences that do not involve overt negation. However, learners can distinguish grammaticality based on the semantics of the licensing verb or adverb, even if judgements are markedly less robust than the taught cases. This leads GMW to suggest that instruction may have a facilitative effect for EFL learners.

For L1 Arabic and L1 Chinese learners of English, a facilitative effect would also be expected from L1 transfer. The corresponding Najdi Arabic ʔayy and Chinese renhe/wh-existentials have a similar NPI distribution to any. The results provide limited grounds for assuming general positive transfer. There may be specific L1 effects for the L1 Chinese-speaking learners, who demonstrate a greater difference in acceptance of negative verbs versus negative adverbs as licensors (average accuracy 2.23 versus 3.36, respectively, on grammatical conditions). By contrast, the L1 Arabic group does not discriminate (2.88 grammatical negative verb versus 2.92 grammatical negative adverb). Gil et al. () suggest that the form of Chinese negative adverbs may facilitate this distinction as the negator bu is often incorporated in these adverbs, e.g., jihu bualmost not ‘hardly’. The implicit negative meaning of such English adverbs is rendered more transparent for L1 Chinese learners given the association with an overt negator in the corresponding L1 lexical items.

A final relevant finding is that at least some learners could discriminate consistently between the different conditions and so seemed to have completely acquired the relevant linguistic properties of NPI any and its licensing conditions. Individual accuracy was calculated on the basis of rejecting 3 out of 4 tokens in each of the ungrammatical conditions and accepting 3 out of 4 in the grammatical conditions. On this measure, 9 of 22 L1 Chinese participants could consistently and accurately distinguish appropriate versus inappropriate syntactic contexts for any, and 10 of 25 advanced-proficiency L1 Arabic speakers.

1.3 Properties of German

The similarities between English, Arabic and Chinese with respect to the distribution of existential NPI any may facilitate knowledge of NPI distribution. The overall patterns of judgements by L1 Arabic and L1 Chinese-speaking learners do not bear this out in detail. Learners demonstrate much less robust knowledge in environments involving lexical-semantic negation from verbs or adverbs licensing any. Given that Chinese and Arabic pattern similarly to English in pertinent respects, an interesting exploration of L1 effects would come from how learners acquire the distribution of any when the L1 provides no facilitative role. This is the case with German.

The closest equivalent to any is represented by forms with the indefinite marker irgend-. The irgend-series shares a number of functions with English any (see ). This overlap motivates the assumption that the learner grammar will map irgend- onto any in a Feature Reassembly model (). However, the syntax and semantics of the items differ in some subtle ways, potentially leading to learnability issues. Most pertinent as motivation for this study is the fact that the irgend-series is not subject to NPI distributional requirements. As we will see later in the discussion, clarifying the specific pattern of results requires a more detailed study of specific interpretive properties. We return to this below as it is necessarily somewhat speculative given that this was not included as a variable in the replication study.

The irgend-series is formed by prefixing irgend to existential or indefinite phrases (wh-words, pronouns such as jemand someone or determiners such as ein- [=a]). Where there is optionality between a non-prefixed existential and an irgend-compound, irgend- introduces an implication of ignorance or indifference on the part of the speaker as in (6) (see ; ).

(6)a. Hier habe ich was gefunden.
here have I what found
‘I have found something here.’
 b. Hier habe ich irgendwas gefunden.
here have I irgend.what found.
‘I have found something or other here’.

While irgend-forms are not NPIs (as evidenced by 6), syntactic context does affect their interpretation as free choice any(thing) versus non-specific some(thing) (see ). Sentences 7–10, which translate the ungrammatical English sentences in Table 1, illustrate these interpretations. In each case, an irgend-form is possible, but will be interpreted as non-specific something/one rather than as an NPI.

(7) Ich habe schon irgendwelche Neuigkeiten gehört.
I have already irgend.which new things heard
‘I have heard some news or other.’
(cf. *I have heard any news)

(8) Irgendwer ist den Anleitungen nicht gefolgt.
irgend.who is the instructions not followed
‘Someone or other did not follow the instructions’
(cf. *Anyone did not follow the instructions)

(9) Ich nehme an, du weißt schon irgendwas über meinen Besuch.
I take PRT you know already irgend.what over my visit
‘I guess that you’ve heard something or other about my visit.’
(cf. *I guess that you know anything about my visit.)

(10) Lucy hat wahrscheinlich letzte Woche irgendwas gekauft.
Lucy has probably last week irgend.what bought.
‘Lucy probably bought something or other last week.’
(cf. *Lucy probably bought anything last week)

The irgend-series also occurs in the grammatical contexts for NPI any, illustrated in (11)–(15). A slight complication comes from the fact that German has a negative quantifier kein- which expresses not-any. An irgend- compound under the sentential negator is not possible without a specific pragmatic intention. For the sake of completeness, example (13) illustrates this pattern. The implication in this sentence is that the homework assignment is something special, i.e. “not just any old homework”.

(11) Hast du irgendwelche Hausaufagaben?
Have you irgend.which homeworks?
‘Do you have any homework (at all)?’

(12) Der Lehrer hat keine Hausaufgaben erteilt.
The teacher has no homeworks distributed
‘The teacher didn’t set any homework’

(13) Die Lehrerin hat nicht irgendwelche Hausaufgaben erteilt.
the teacher has not irgend.which homeworks distributed.
‘The teacher didn’t set just any old homework.’

(14) Ich bedauere, daß ich irgendwas über deine Fahrprüfung gesagt habe.
I regret that I irgend.what over your driving test said have.
‘I’m sorry I said anything (at all) about your driving test.’

(15) James hat bei der Feier kaum irgendwas gegessen.
James has at the part hardly irgend.what eaten.
‘James hardly ate anything at all at the party.’

Let’s return to questions of learnability that this comparison raises. As already noted, Feature Reassembly () would assume that L1 German-L2 English learners map the lexical item any onto the features associated with irgend-, due to the shared functions and distribution. However, one element of the subsequent learning task is to restrict the distribution of any as an NPI. In all contexts where any occurs, a representation which implicates the existential indefinite properties mapped from irgend- would grammatically license the occurrence, but potentially assign a different interpretation (see above). Crucially from a learnability perspective, ungrammatical usages would also be licensed by this representation, and on the basis of positive evidence, restructuring poses a problem. The presence of the some/any distinction in English might aid in arriving at a target representation. Nevertheless, without negative evidence, it is difficult to rule out ungrammatical occurrences of any in L1 German-L2 English. In many contexts, there is optionality between some and any, complicating the task of arriving at a target NPI representation which specifically disallows any in certain contexts. In a judgement task, where ungrammatical occurrences are presented, learners who have not reassembled features so as to restrict the distribution will syntactically license the usage (though potentially with different meanings).

Of course, as noted in the studies of teaching materials by GMW, negative evidence is available from instruction to the effect that any is restricted to overtly negative contexts or questions as compared to some (see for similar findings from EFL materials for German-speaking learners). In common with GMW’s predictions, the signature of reliance on a learned pedagogical rule would be a rejection of grammatical sentences where licensing of any is due to semantic features of verbs or adverbs, and acceptance of ungrammatical sentences with any outside the scope of negation.

Based on the discussion so far, and the insights furnished by GMW’s experiments, we explore potential effects of pedagogical rules in explicit input, versus observability in incidental input. The hypothesised interplay between these factors and the eight sentence types is outlined in Table 2.

Table 2

Interplay between potential sources of knowledge of any and the eight sentence types.




2Negative declarativegrammatical++

3Negative verbgrammatical+

4Negative adverbgrammatical+

5Affirmative declarativesungrammatical+

6Declarative outside scopeungrammatical

7Non-factive verbungrammatical

8Possibility adverbungrammatical

Note: – = a particular source of knowledge is not associated with a sentence type. + = a particular source of knowledge is associated with a sentence type.

As discussed in section 1.2, pedagogical rules focus strongly on occurrences in questions and negative declaratives, facilitating knowledge of sentence types 1, 2, and – as a common pedagogical counter example – sentence type 5. Sentences involving any licensed by semantically negative verbs and adverbs occur in incidental input. Knowledge of these is therefore in principle derivable from positive evidence, even though these features are not typically taught. Ungrammatical occurrences of types 6–8 in Table 2 are not taught. They are by definition not available from incidental input as ungrammatical sentences do not occur, and they would not be ruled out by mapping from L1 German.

Key comparisons are therefore (i) between the taught grammatical questions and negative declaratives versus untaught types with negative verb and adverbs. A difference in the acceptability ratings would index potential effects of instruction. Such instructional effects can be further explored by (ii) comparing the rating accuracy of grammatical questions (1) versus parallel ungrammatical affirmative declaratives (5), and also by contrasting negative declaratives (2) with sentences containing scope violations (6). Because the ungrammatical sentences are by definition absent from positive input, teaching effects would be indexed by more accurate ratings of sentence type 1 versus 5 and 2 versus 6.

Finally, input and L1 effects would be identified by comparisons of (iii) grammatical, observable but untaught sentence types with negative verbs (3) and adverbs (4) versus ungrammatical ones with non-factive verbs (7) and possibility adverbs (8), which are neither taught nor observable. A tendency to reject ungrammatical sentences altogether would indicate an overgeneralisation of the pedagogical rule, while a tendency to accept all of them would indicate L1 effects. More accurate ratings in sentence types 3 and 4 against 7 and 8 would furthermore point towards developing grammatical L2 knowledge despite the absence of an overt licensor.

2. The Study

2.1 Participants

72 learners of English studying in an English Language Teaching degree programme participated in the study. This sample size would permit detection of effects as small as ηp2 = 0.13 (; ). Participants’ mean age was 22.48 years, (SD = 3.29), and German was their L1, either mono- or multilingually. 53 identified as female, 18 as male, and one as non-binary. All participants had had eight years of regular, instructed EFL school teaching prior to their tertiary studies; in addition to that, around 11% reported a stay abroad between one and six months, while around 14% had stayed for a longer period of time in an English-speaking country. The great majority, though, had never had an extended stay in an English-speaking environment (75%).

All 72 participants took part voluntarily, anonymously, and with explicit consent, but without any financial renumeration. At the time of testing, participants had been studying English at tertiary level for at least two years and were approaching C1 proficiency level (), i.e. “advanced” as defined by the Common European Framework of reference for language. This was measured by proficiency tests at the beginning of the degree and after two years in the study programme. The data of all 72 participants were used for reliability and bias analyses of the acceptability judgement ratings.

2.2 Materials and Method

As in the GMW studies, participants rated the naturalness of sentences containing any in a paced acceptability judgement task. 12 lexicalisations for each of the sentence types were created, resulting in 96 tokens, which were distributed in pseudo-randomised experimental lists following a Latin Square design. Participants saw 24 randomised test items (3 items per experimental condition) as well as eight fillers, either systematic distractors or items piloting a separate study on negative inversion. Test items, raw data and R-script are accessible via https://osf.io/9qt46/.

The acceptability judgement task was administered online through SoSciSurvey (). Participants were informed that the experiment was part of a study about learners’ grammatical knowledge, the procedure was explained briefly, and two untimed items for practice were provided.

The sentences were rated on a four-point Likert scale, labelled I’m sure this sounds natural, I think this sounds natural, I think this does not sound natural, and I’m sure this does not sound natural. Since the acceptability judgement scores were intended to elicit participants’ intuition about the naturalness of a given sentence, a gradable rating option was chosen instead of a binary right-wrong choice. Contrary to Marsden et al. (), an even point-scale was chosen so that participants were forced into one or the other direction, thereby avoiding clusters in the classic middle-ground option cannot decide. For the same reason, the option I do not know was omitted. Thus, it was assumed that even mild and subtle intuitions about a sentence’s acceptability could be captured.

2.3 Data Coding and Analysis

Signal Detection Theory (SDT) was used to analyse how participants discriminated between grammatical and ungrammatical sentence types and whether they showed any general bias towards accepting or rejecting (). Huang & Ferreira () propose using SDT for judgement data as it delivers insights into how well participants differentiate between grammatical and ungrammatical sentences and whether they have any bias in their preferences. To this end, acceptability ratings were coded as follows. Detecting a grammatical sentence’s grammaticality resulted in a hit; rejecting a grammatical sentence resulted in a miss. If participants correctly detected the ungrammaticality of a sentence, they scored a correct rejection, or true negative. Where they rated ungrammatical sentences as grammatical, they scored a false alarm. Based on this 2×2 matrix, the SDT indices d’ and c were calculated; d’, a measure of sensitivity, captures participants’ reliability when distinguishing between grammatical and ungrammatical sentences. Index c, the criterion location index, expresses conservative or liberal biases.

Mixed ordinal regression modelled potential effects of sentence type on the acceptability ratings. The four Likert points were treatment-coded as ordered categorical data, with the levels “1”, “2”, “3”, and “4”, “1” reflecting the least appropriate, and “4” reflecting the most appropriate rating. Thus, a rating such as I’m sure this sounds natural for a grammatical sentence would be coded as “4”, most appropriate, while the same rating for an ungrammatical sentence would score a “1”, least appropriate. The sentences were sum-contrast-coded, so that the predictor levels represent the difference between these levels and the grand mean, thus effectively ‘centring’ the effects of the variable’s levels at this value (). This coding was chosen since none of the eight sentence types could reasonably be thought of as a baseline level against which the other estimates would then be calculated.

For the mixed modelling, we used cumulative link mixed regression with Laplace approximation from the ordinal package () in R (). Main regression effects from the cumulative link mixed model are reported based on Type-II Wald χ2-tests, as implemented in the RVAideMemoire package (). Partial effects statistics (Wald) were derived from the model summaries produced within ordinal.

3. Results

After replacing missing values by appropriate imputations, participants’ rating accuracy in both the grammatical and ungrammatical sentences were quantified. The heavily left-skewed (Fisher-Pearson g1 = –1.30) overall distribution towards category “4” (most appropriate) could be interpreted as a first indication of above-average accuracy. In order to examine if accuracy holds across grammatical and ungrammatical sentence types, the ratings were coded as illustrated in Table 3.

Table 3

Contingency table for the counts of the four SDT coding and resulting counts.



ratingsacceptable693 hits 134 false alarms

unacceptable150 misses 692 correct rejections

Table 3 is almost perfectly symmetrical, with about the same number of true positives (hits and correct rejections). This is a first indication of participants’ appropriate sensitivity towards grammaticality. Accordingly, a d’ value of 1.91 (SE = 0.01, 95% CI [1.90, 1.92], significant on an inter-individual level (t(71) = 22.42, p < .001), shows that the acceptability ratings are 40% away from chance level, which in turn suggests a rather reliable discrimination between grammatical and ungrammatical sentence types. In line with these results, there was hardly any bias towards one particular type of response (acceptable, unacceptable) in the participants’ ratings either (c-value = .03, t(71) = 0.99, p = 0.32); in other words, the learners were neither particularly conservative nor liberal in their acceptability ratings.

Consider Figure 1; it visualises the conditional means for the participants’ acceptability judgement ratings across the eight sentence types.

Figure 1 

Conditional means plot for acceptability judgement ratings by sentence type.

Note: The dashed horizontal line represents the grand mean. The vertical arrows illustrate the distance between each mean sentence type rating and this grand mean. Error bars illustrate 95% confidence intervals.

There are pronounced differences in conditional means; sentence types 1, 2, 4, and 6 elicited ratings above the grand mean, while others are below, with sentence type 3 deviating most markedly. The conditional means across all eight sentence types are significant (Type II Wald ANOVA, χ2(7) = 91.77, p < .001), and differences in individual repeated measure ratings were modelled using a cumulative link mixed model with Laplace approximation. Random intercepts and slopes for participants were included, as their variance was substantial (S2 = 0.21, SD = 0.47); item variance, in contrast, was minute (values < .001). The model’s fixed and random effects are illustrated in Table 4.

Table 4

Coefficients from a cumulative link mixed model (Laplace approximation) with random intercepts and slopes for participants.



fixed effects

   threshold coefficients

     1 | 2–3.260.04–1.800.15–<.001***

     2 | 3–2.120.12–1.170.12–17.640.090.15<.001***

     3 | 4–0.480.62–0.260.1–4.670.510.76<.001***



     negative declaratives0.962.610.530.224.431.714<.001***

     negative verbs–1.520.22–0.840.2–7.480.150.33<.001***

     negative adverbs0.

     affirm. declaratives–0.310.73–0.170.2–1.580.51.08.11

     negative declarative0.421.520.<.05*

     non-factive verb–0.690.5–0.380.16–4.240.370.69<.001***

     possibility adverb–0.210.81–0.120.17–1.210.541.220.29

random effectsvarianceSDcorr.


     negative declaratives0.470.69–0.52

     negative verbs1.741.32–0.600.64

     negative adverbs0.410.64–0.490.870.60

     affirm. declaratives1.661.29–0.40–0.18–0.300.41

     negative declarative1.711.31–0.730.410.480.100.66

     non-factive verb1.541.24–0.550.200.12–0.150.870.93

     possibility adverb1.231.11–0.440.08–0.05–0.270.920.850.99

Note: Number of partial effects = 8, number of observations = 1560, total N = 65, CI = confidence interval, LL = lower limit, UL = upper limit, OR = odds ratios, pseudo-R2 (McFadden) = 0.09, Cox and Snell (ML) = 0.17, Cox & Snell’s R squared () = 0.19. a Cohen’s d effect size approximations were calculated using (log(OR)x√3)/π, . Condition number of the Hessian, measuring the empirical identifiability of the model, is 0.02. Maximum absolute gradient of the log-likelihood function with respect to the parameters is 0.34.

Table 4 shows that five out of the eight sentence types significantly influence the acceptability ratings. Medium to large effect sizes (d ≥ 0.5, ), however, can be seen for sentence types 1, 2, and 3 only. Here, questions and declaratives show odds ratios greater than 1, while negative verbs show a ratio below 1. The odds ratio of 0.22 for sentences containing negative verbs, for instance, means that the odds of getting a more accurate acceptability judgement rating is 0.22 times that of the grand mean, or 78% lower than with sentences on average, holding all other sentence types constant; in other words, this model would predict that participants rate this sentence type much less accurately than they do on average. While this model suffers from minor scale effects, as the proportional odds assumption does not hold perfectly across all predictor levels, the estimates still provide a useful and reliable measurement of the sentence type effects (; ).

So far, analyses have shown that participants can reliably, and in an unbiased fashion, discriminate between grammatical and ungrammatical sentences. Also, the accuracy of participants’ ratings clearly depends on licensing conditions for any; some sentence types trigger better, some worse ratings than on average. Particular difficulties seem to arise with sentences containing negative verbs. This is noteworthy: learners are more reliably able to reject ungrammatical any embedded under non-factive verbs than they are able to accept grammatical occurrences licensed by negative verbs. Indeed, grammatical use of any licensed by negative verbs produces the worst ratings of all sentence types.

We then examined the influence of potential sources of knowledge, as outlined in Table 2. Recall that this comprised the five sentence-pair and pooled sentence type contrasts 1+2 versus 3+4, 1 versus 5, 2 versus 6, 3+4 versus 7+8, and 1–4 versus 5–8. In addition, the contrast between sentence type 3 and 7 was explored, comparing grammatical negative verbs with ungrammatical non-factive verbs. Table 5 and Figure 2 illustrate the effects for each of these contrasts, based on monofactorial cumulative link mixed models with random intercepts and slopes for participants. Significant pairs in Table 5 are also mirrored in estimated marginal means contrasts (Tukey-adjusted p-values).

Table 5

Main and partial fixed effects for six sentence pairs and pooled sentence contrasts.



1+2 vs 3+4taught – untaughtχ2(1) = 65.970.810.098.87<.001***

1 vs 5questions – affirmative decl.χ2(1) = 26.420.750.154.90<.001***

3 vs 7neg. verbs – non-factive verbsχ2(1) = 8.47–0.440.15–2.95<.01**

2 vs 6neg. decl. – neg. decl. scopeχ2(1) = 2.840.

3+4 vs 7+8licensor – no licensorχ2(1) = 1.340.

1–4 vs 5–8grammatical – ungrammaticalχ2(1) =

Figure 2 

Conditional means plots for six sentence pairs and pooled sentence contrasts.

As shown in Table 5 and Figure 2, significant contrasts arise between taught and untaught sentence types, questions versus affirmative declaratives, and sentences containing negative versus non-factive verbs. In the top-left panel of Figure 2, the significant positive slope towards taught sentences shows that participants produced significantly more accurate acceptability judgement ratings for sentences representing typically taught properties compared to the untaught properties. In the top middle panel, we can see that questions elicited significantly more accurate ratings than affirmative declaratives, which might indicate an effect of instruction given that questions are taught as a prototypical grammatical context for the use of any. The difference between sentences involving negative and non-factive verbs, visible in the top-right panel, is, again, surprising. While this property is not generally addressed in instruction, learners can encounter the licensing of any from negative verbs in positive input. However, this type of grammatical sentence received a significantly less accurate rating than the ungrammatical condition with non-factive verbs. Such an effect cannot be observed for the two adverb conditions (sentence types 4 and 8), indicating that learners were reliably able to accept grammatical sentences and reject ungrammatical counterparts in the face of a similar acquisition task as that involving verbs. As we can see in the bottom panels in Figure 2, scope, the absence of a licensor in grammatical and ungrammatical sentences as well as grammaticality in general do not significantly influence rating accuracy.

4. Discussion and Conclusions

Overall, the results indicate an ability to distinguish acceptability of NPI contexts for the occurrence of any by advanced L1 German-speaking learners of English, with a complication related to judgements of licensing from verb semantics. The discussion considers the potential sources of knowledge and interaction between them in leading to this pattern.

It can be stated that, by this level of proficiency, learner knowledge is not generally conditioned by L1 properties. Recall that the German irgend-series would be mapped to any in a Feature Reassembly framework and that this mapping would license the occurrence of any in all of the sentence types tested, meaning that L1 influence would be indexed by a tendency to accept ungrammatical sentences. Clearly, the learners in the study have restructured grammars that permit broad discrimination of licensing conditions. A question is the relative contribution of pedagogical and incidental input to this restructuring. Taking pedagogical input first, it seems clear that explicit pedagogical rules can have a facilitative effect, supporting GMW’s results. Those licensing contexts typically addressed in pedagogical grammar evince significantly more consistent target judgements than conditions which are not addressed in instruction.

The other side of this coin is the question of untaught and (un)observable properties. Even if performance on the untaught properties is less robust, effects of observability would be evidenced by an ability to distinguish grammaticality on the basis of the semantics of licensing verbs or adverbs. In this area, the results from the present study diverge from the previous findings, as illustrated in Table 6.

Table 6

Mean acceptability judgement ratings in all eight sentence types across four experimental studies.



L1 ARABIC (N= 25)L1 CHINESE (N = 22)L1 ENGLISH (N = 15)L1 GERMAN (N = 65)

Sentence typesQ/DeclGrammatical3.843.863.933.74


Neg ScopeGrammatical3.683.914.003.71


Main VerbGrammatical2.882.233.732.70




Unlike previous studies, our learners produced considerably more accurate ratings for ungrammatical any embedded under non-factive verbs. As well as diverging from previous learner groups’ results, this stands out as distinct from the native speakers in GMW’s studies. For all these groups, grammatical sentences consistently evince more accurate judgements than ungrammatical sentences, but this pattern is reversed for the L1 German group’s performance on the verb-licensing condition. Focussing only on the learners, performance on grammatical sentences with negative verbs is numerically similar (2.88 L1 Chinese, 2.70 L1 German, 2.23 L1 Arabic). The outlier that calls for explanation is specifically the better-than-expected performance in rejecting ungrammatical sentences. This result is puzzling and is not conducive to a straightforward analysis based on the learnability issues and L1 properties discussed above.

We finish by offering some speculation on a possible analysis. This assumes the specific pattern of results is due to a confluence of factors involving complications in the interpretation of any licensed by verbs, compounded by L1 effects. This points to avenues for future research. It can be noted that while the German equivalents of the English main-verb conditions are grammatically possible, there is a subtle pragmatic/semantic distinction, which impinges upon the acceptability of the existential with the different types of matrix verbs. This is illustrated in (16) and (17), which extend example (4) from earlier.

(16)Maria bedauert, dass sie irgendwas bei der Feier gegessen hat.
Maria regrets that she irgend.what at the party eaten has.
‘Maria regrets that she ate anything/something at the party.’

(17)Maria glaubt, dass sie irgendwas bei der Feier gegessen hat.
Maria thinks that she irgend.what at the party eaten has.
‘Maria thinks that she ate something at the party.’

In (16), the irgend-existential can be interpreted as meaning that Maria did in fact eat something, but regrets it for example because she got food poisoning or was intending to diet that day, etc. Or it can mean that she regrets eating anything at all. These readings are pragmatically unexceptional and available for a sentence in isolation. By contrast, the most natural reading of (17), as indicated in the translation is as a non-specific some-existential, with the interpretation that Maria is unsure whether or not she actually ate some foodstuff. Remember that irgend- also functions to introduce an implication of uncertainty or ignorance. Such a reading for a sentence in isolation is obviously somewhat pragmatically odd. If there is in fact some level of L1 influence, resulting in this pragmatically odd reading, this may explain the more reliable rejection of the ungrammatical English sentences. In other words, the learners have acquired the basic NPI distributional patterns, but may still be influenced by L1 properties in the interpretation of existential any in other contexts, giving rise to free-choice or indefinite meanings. Of course, this raises the question of why the verb conditions pattern differently to the adverb conditions, which are also not the core NPI contexts. Grammatical adverb conditions would receive an NPI reading in German while ungrammatical conditions would also receive a non-specific some reading (see 18 and 19). However, in these cases, there are no additional complications with respect to pragmatic interpretation. So, if the idea that there is subtle L1 influence is on the right lines, we would not necessarily expect a difference to emerge in these conditions.

(18)Maria hat kaum irgendwas bei der Feier gegessen.
Maria has hardly irgend.what at the party eaten
‘Maria hardly ate anything at the party.’

(19)Maria hat wahrscheinlich irgendwas bei der Feier gegessen.
Maria has probably irgend.what at the party eaten.
‘Maria probably ate something or other at the party.’

Furthermore, judgements involving adverbs are generally better than the verb conditions across the board (see Table 6). It may be that there is something about adverbs that facilitates acquisition of NPI licensing prior to licensing from verbs. This something might be frequency if the adverb contexts occur more frequently in input. It might also be another indirect pedagogical effect. Recall that GMW’s survey of teaching materials found that pedagogical rules did sometimes mention co-occurrence of any with specific adverbs, even if this was often presented as an ‘exception’ to the rule. Realistically, it may be a confluence of this range of factors which leads to the differences.

The GMW studies, and by extension this replication, only focussed on NPI distribution of any. It is clear from the results from L1 German speakers learning English that future research should explore comprehension and pragmatics in order to gain a fuller picture of how any is acquired and whether NPI and indefinite or free-choice meanings and pragmatics are implicated. Especially for L1 German-speaking learners, results from acceptability may mask a more complex picture related to the semantic and pragmatic readings assigned to different types of sentences, which may implicate continued L1 influence on existentials and indefinites at the level of semantic and pragmatic interpretation.