Gender agreement

The monograph by Nina Sumbatova and Yury Lander (with Magomed Mamaev) “The Dargwa dialect of the village of Tanti. A grammar sketch. Problems of syntax” is based on the field data obtained in 2008–2013 in the village of Tanti of the Akusha district, Republic of Daghestan (Russian Federation). The dialect of Tanti belongs to the Tsudakhar group of the Dargwa dialects. It has not been described before.
The book by Nina Sumbatova and Yury Lander contains a grammar sketch of Tanti (Part 1) and a more detailed description of some syntactic problems – NP structure; verbal constructions; the structure of different sentence types (declarative, interrogative, etc.) with different types of information structure, as well as agreement in noun class and person (Parts 2–6). The book also provides a comparative description of person agreement in the Nakh-Daghestanian language family.
Except the grammatical description, the book contains a small corpus of oral texts in the dialects of Tanti (11 texts, 55 pages) and a glossary of about 1300 lexical entries.

Монография Н.Р Сумбатовой и Ю.А. Ландера (при участии М.Х. Мамаева) «Даргинский говор селения Танты. Грамматический очерк. Вопросы синтаксиса» (общий объем монографии – более 32 а. л.) основана на полевых исследованиях авторов, осуществлявшихся в 2008–2013 гг. в селении Танты Акушинского района Республики Дагестан. Тантынский идиом относится к цудахарской группе даргинских диалектов. Его грамматических описаний ранее не существовало.
В книге содержится грамматический очерк тантынского идиома (Глава 1), а также подробное описание ряда разделов синтаксиса – структуры именной группы, глагольных конструкций, структуры простых предложений различных типов (повествовательных/ вопросительных и т. д.) и с различными видами коммуникативной структуры, а также категории согласования по именному классу и лицу (Главы 2–6). Кроме того, содержится сравнительное описание личного согласования в тех языках нахско-дагестанской семьи, где оно представлено.
Помимо грамматических разделов, книга содержит небольшой корпус глоссированных текстов на тантынском диалекте (11 текстов, объем около 2,4 а.л.) и словник тантынского диалекта (около 1300 лексем).

The language learning process by third age learners has recently become a field of interest for linguistic research. However, it has been studied mostly regarding psychological and cognitive considerations (stress, motivation, attention, memory, etc.), rather than according to language internal factors.
The present study aims to provide a contribution by investigating the mastery of Italian grammatical gender by third-age Swedish native speakers who are learning Italian as an L2.
The study wants to verify if there are any particular patterns in grammatical gender learning according to the age of the learners, with a focus on third age, especially regarding formal cues (such as phonology, morphology, syntax) vs. semantics.
Our research shows that elderly learners seem to be sensitive to semantics, while younger learners seem to be more sensitive to phonology and syntax; psychological factors seem not to play an important role.
Moreover, contrarily to our hypotheses, we found that elderly informants show a much greater homogeneity in their results, in comparison to young informants, and a greater ability in problem solving.
Another interesting finding is that, in general, elderly informants outperform younger ones who had spent an equal amount of time attending Italian lessons.
In sum: psychological explanations do not seem to justify an age differentiation in gender learning, while language internal factors seem to be at the basis of a general, age-dependent, differentiation.

The article focuses on some characteristics of gender marking in the case of professional designations in Russian and compares them with their Slovenian equivalents.
In Russian sentences with attributive meaning, the use of masculine forms in subject complement is more frequent than in Slovenian.
Since Russian feminatives are less productive than Slovenian ones, in some professional
designations Russian uses “hybrid nouns” that can refer to both genders despite belonging to the first declension. The female referent in this case can be marked by analytical forms of the predicate (Врач пришла 'The doctor arrived'). This type of agreement is less frequent in the case of noun phrases with modifiers (наша врач 'our doctor', хорошая врач 'a good doctor'), where, though the referent is a woman, the masculine form is preferred.
In Slovenian, the referential gender is most commonly marked by feminine derived nouns, so its marking with analytical predicate forms is not typical. The only exception is the noun vodja 'leader', which lacks a feminative, and for that reason licenses the gender marking of the referential gender with its syntactical targets—predicate verbs (Obe vodji sta prišli 'Both leaders arrived') as well as modifiers within the framework of noun phrase (naša vodja 'our leader', umetniška vodja 'art director'). In contrast to Russian equivalents, the feminine forms of the modifier are used in non-nominative cases as well (Projekt smo oddali umetniški vodji 'The project was submitted to the art director').

Grammatical gender is claimed to be easily acquired in Romance languages, especially in Italian and Spanish. This can be observed in the early manifestations of gender agreement and in the low rate of agreement errors. In contrast, German gender is acquired relatively late, and children make more errors. Previous studies
with data from our project have shown that, despite the easy acquisition of the Spanish gender system, bilinguals differ from monolinguals. Fillers and protoarticles are less frequent in the Spanish bilingual utterances; this concerns the first utterances until the age of around 2 years (Lleó 2003, Kuchenbrandt 2005). The lower frequency of nouns with (proto-)determiners can be attributed to the delayed acquisition of the required prosodic structures, which is independently attested in the bilingual Spanish data (Lleó 2002). In this study, we want to contrast the two languages, and we will address the question whether we find a
Spanish influence on German as well. Following Kupisch (2006), the acquisition of Romance determiners by a bilingual may bootstrap the acquisition of determiners in his Germanic language, leading to an earlier convergence to the target system. In previous research, no difference has been found between monolinguals
and bilinguals with respect to the acquisition of higher prosodic domains and the development of (proto-)articles in German (Lleó 2002). However, we have not yet focused directly on the development of determiners. This study shows a slight accelerating effect in bilingual German.

Grammatical gender processing is a classic topic in the neurolinguistic literature. However, most of the studies consider major standard languages, such as English, Dutch, Spanish or Italian. This is a limitation, since these languages display relatively simple systems that are fairly stable and do not allow to address language variation and change. Here we investigated gender agreement in a dialect of Southern Italy spoken in Agnone, using Event-Related Potentials. Compared with Italian and most standard Romance languages, the Agnonese gender system is more complex (with three target genders: feminine, masculine, neuter) and, also under the pressure of standard Italian, undergoing change, with the neuter merging into the masculine. Results showed that all determiner-noun gender mismatches elicited robust P600 and moderate Left Anterior Negativity (LAN) effects. Yet not all violations involving neuter nouns were equally outright. For masculine-neuter, the amplitude of the P600 was less pronounced than for feminine-neuter violations. Moreover, when looking at individual differences, the P600 for masculine-neuter was progressively reduced as participants' competence in the Agnonese conservative variety decreased. Since sociolinguistics shows that individual variation within the speech community may reflect linguistic change, the smaller P600 exhibited by less conservative speakers could be described as a brain signature of ongoing language change in the Agnonese gender system. In other words, the attenuation of ungrammaticality-driven brain responses may provide a measurable indicator of the fading of the neuter and its merging into the masculine. This work paves the way to the neurolinguistic description of non-standard varieties and also breaks ground for a "diachronic neurolinguistics", extending investigation into the neurobiology of language beyond the syn-chronic dimension.

Celtic Linguistics Conference, Cardiff University, September 2016

The received view in Manx scholarship is that grammatical gender in Manx ‘has been rather badly confused’ (O’Rahilly 1972: 119). Broderick (1999: 165) and Thomson (1986: 9) both claim that the feminine gender in inanimate nouns was largely defunct and replaced by the masculine even in seventeenth or eighteenth century Manx.

In this paper evidence will be adduced to show that in fact the grammatical gender system in Manx was maintained intact, much as it is in Irish and Scottish Gaelic, at all periods of Manx from the seventeenth to the nineteenth century, with the exception of the terminal speakers recorded in the twentieth century. Instances of gender concord / lack of concord in terms of lenition and pronoun replacement were counted in five texts from different periods of the language, plus the transcriptions of two of the terminal speakers (Broderick 1984–6 I). In all five texts occurrence of expected lenition with feminine inanimate nouns was between 70.9% and 98.2%, while occurrence of expected pronoun replacement was between 54.0% and 83.3%.

The most striking result was that in the writings of Edward Faragher (b. 1831) expected lenition is found in 93.6% of cases and expected pronoun replacement in 83.3%, while in the speech of his younger neighbour and acquaintance Ned Maddrell (b. 1878) the figures are 35.7% and 0.0% respectively. To some extent this could reflect the difference between careful writing and spontaneous speech, but overall, it is likely that Faragher’s usage reflects the maintenance of grammatical gender in his grammar while Maddrell (and the other terminal speakers) have not acquired grammatical genders because of inadequate exposure and inadequate socialization in the language. Loss of grammatical gender is thus not a feature of gradual change over centuries but a sudden development in the final generation of speakers as a result of language shift.

The case of Maddrell, who acquired English from his parents before starting to acquire Manx from an elderly relative between the ages of three and five, is compared with the case of Greek-English bilinguals studied by Unsworth et al. (2012). Based on such research, it seems likely that the overall quantity of input for the terminal speakers of Manx was not sufficient for full acquisition of redundant features such as gender. Late onset of acquisition of Manx, and the existence of an already developing linguistic system lacking grammatical gender (English), in the case of Maddrell, could also impede the acquisition of the gender system. In contrast, earlier generations of speakers growing up in a much more Manx-dominant community would have no such problem. It is likely that the received view argued against here is rooted in an assumption that Manx is a “decayed” variety of Gaelic, represented by O’Rahilly’s (1972: 121) well-known claim that ‘[f]rom the beginning of its career as a written language English influence played havoc with its syntax… Manx hardly deserved to live’.

This study presents data from an elicitation study on French size and color adjectives in noun phrases (DPs), both early acquired structures. Thirty-two francophone children aged 3 – 5 years participated in the study. Adjectives were elicited using specially designed puzzles and spontaneous speech corpora. We observed that errorsin French variable adjectives are produced in the early acquisition stages, especially in the context of feminine colour DPs. We propose that the source of difficulty for feminine variable adjectives is the retrieval of a lexicalized form that competes with the masculine adjective denoting the same concept.

The goal of this paper is to compare the learning outcomes of beginning online students (n=14) of Spanish with classroom students engaged in a hybrid or blended course (n=19). All students were native speakers of English. Some studies show that online students perform better than classroom students, while other studies show no significant differences. All previous studies use general measurements of proficiency or scoring instead of examining particular differences within specific linguistic elements. This paper focuses on gender agreement and gathers acquisitionist research on the matter to weight the factors that affect its learning in online and classroom students. We extracted 2777 tokens from writing tasks and performed multivariate analysis using Rbrul (Johnson) to test gender accuracy against delivery mode and linguistic factors such as animacy, morphological endings and gender assignment of the noun, as well as the grammatical category of the agreeing word. Results suggest that the delivery mode does not play a role in the overall performance of online versus classroom students. However, classroom students perform better with animate nouns, while online students do so in a non-significant manner. Physical presence of people may favor the learning of gender agreement with animate referents.

An analysis of 519 gender errors (out of 9,378 modifiers) in the advanced French interlanguage of 27 Dutch L1 speakers confirms earlier findings that gender assignment and/or agreement remain problematic for learners at all levels. A hypothesis derived from Pienemann's Processability Theory (1998a) that accuracy rates would be higher for gender agreement in structures involving no exchange of grammatical information between constituents was not confirmed. The analysis of interindividual and intra-individual variation in gender accuracy rates revealed effects from avoidance and generalisation strategies, from linguistic variables, sociobiographical variables and psycholinguistic variables. We argue that gender errors can originate at the lemma level, at the gender node level, or at the lexeme level. Different psycholinguistic scenarios are presented to account for intra-individual variation in gender assignment and agreement.

A comparative literature review is carried out on the topic of pronominal reference, looking specifically at pronoun gender agreement. Theoretical (Audring 2013, Josefsson 2006) and empirical accounts (Hammer et al. 2005, Dong et al. 2015) are compared. It is argued that these converge on the finding that pronouns, rather than being mere syntactic devices used for agreement, are selected on the basis of their own semantic load (referred to here as “independent pronoun semantics”); moreover, that independent pronoun semantics plays an important role in establishing pronominal reference. These findings are placed in a broader context, making a case for the psychological reality of Gricean pragmatics. Finally, a connection is made with Everett’s model of “cultural constraints on grammar” (cf. Everett 2005).

Term paper for The Syntax-Semantics Interface class, taught by Prof. Dr. Helen de Hoop.

If you find any ERRATA, please let me know!

The paper deals with the functioning of the Ukrainian language among the young generation of Transcarpathian Hungarians, the national minority of this region. Examples from the Ukrainian language
will be compared with examples from other Slavic languages, particularly from the Serbian language of Vojvodina Hungarians and from the Slovenian language of Prekmurian Hungarians. The
main goal is to identify the typological convergence.

A cikkben a kárpátaljai magyarok ukrán nyelvéről van szó, elsősorban a fiatal generáció képviselői voltak az adatközlők. Az ukrán nyelvből való példákat összehasonlítom a többi szláv nyelvekből való példákkal, különösen, a vajdasági magyarok szerb nyelvével és a muravidéki magyarok szlovén nyelvével. A fő célom a tipológiai konvergencia meg-
állapítása.

Inflectional morphology causes persistent difficulties for second language (L2) learners (Montrul, Silvina & Kim Potowski. 2007. Command of gender agreement in school-age Spanish-English bilingual children. International Journal of Bilingualism 11(3). 301–328; Montrul, Silvina, Israel de la Fuente, Justin Davidson & Rebecca Foote. 2013. The role of experience in the acquisition and production of diminutives and gender in Spanish: Evidence from L2 learners and heritage speakers. Second Language Research 29(1). 87–118). Learners operate with a default gender value, and overgeneralize the masculine forms of determiners and modifiers (White, Lydia, Elena Valenzuela, Martyna Kozlowska-Macgregor & Ingrid Leung. 2004. Gender and number agreement in nonnative Spanish. Applied Psycholinguistics 25(1). 105–133; Schlig 2003). 111 essays written were collected containing 799 correct uses and 281 errors from Dutch students whose written ability in Spanish is A2 (Common European Framework). The r...

Статья посвящена согласованию по числу в андийском языке,
которое выражается показателем ‑(V)l. Это явление прежде не
описывалось в грамматиках андийских диалектов как единый
(по крайней мере, морфологически) феномен. Статья призвана
заполнить эту лакуну. В ней перечисляются возможные мишени
согласования по числу: зависимые в именной группе, глагольные
словоформы и наречия, — и рассматривается возможный сценарий развития показателя ‑(V)l в андийском языке.

The L2 use among Transcarpathian Hungarians: sociolinguistic and structural analysis.
The paper deals with the language situation among Transcarpathian Hungarians, the use of the Ukrainian and Russian languages in everyday communication. The influence of Slavic languages depends on many variables, e.g. on the type of settlement. In the Ukrainian language I found the influence from Hungarian and Russian.

This article presents results from a study on agreement with conjoined singular noun phrases in Icelandic. A survey was conducted to elicit agreement choices on two agreement targets (predicate adjectives and personal pronouns) with agreement controllers from four individuation levels (human, animal, countable object, uncountable abstract). The aim was to provide an overview of the strategies for determining agreement with conjoined noun phrases in Icelandic, and examine whether two typological hierarchies-the Agreement Hierarchy and the Individuation Hierarchy-can predict the distribution of agreement options. This research identifies and examines five different agreement strategies with conjoined singular noun phrases in Icelandic. The distribution of these patterns in relation to the two hierarchies is evaluated with the aim of discerning what influences agreement choices. Both typological hierarchies are shown to affect the choice of agreement strategies. In addition to discussing semantic resolution, syntactic resolution, and partial agreement in Icelandic-this paper identifies and argues for two types of neuter default agreement. The first is neuter plural, a fixed value for agreement with conjoined noun phrases, and the second is neuter singular: semantic default agreement triggered by referents that are low on the Individuation Hierarchy. This research presents new data that has implications for the understanding of agreement with conjoined noun phrases and the function of neuter in Icelandic and related languages.

This research investigates the influence of L1 properties and proficiency level on the acquisition of gender agreement addressing linear distance as a possible cause of errors. French and English-speaking learners of Spanish participated in the study. Subjects were divided into four different groups considering their L1 (French and English) and their proficiency level (intermediate and advanced). Subjects completed three different tasks: an untimed grammaticality judgment task (UGJT), an elicited oral imitation test (EOI), and an eye-tracking experiment. As for linear distance, each experimental item was distributed across three distance conditions: distance number 1 corresponds to an adjacent adjective pattern, distance number 2 separates the adjective using a copula, and distance number 3 included between three and four words interceding before the copula. Our results show no linear distance effect for either the explicit task or the implicit task.

This paper focuses on the evolution of the Old Indo-Aryan reciprocal pronoun anyo’nya- as well as some related forms, tracing its grammaticalization from the early Vedic period onwards until the beginning of the Middle Indic period. On the basis of a comparison of the history of this formation with similar processes documented in some other Indo-European branches (Greek, Slavic etc.), I uncover some basic mechanisms and scenarios of the evolution of reciprocal
constructions attested in the history of Indo-Aryan languages in a diachronic typological context, offering a number of typological generalizations on the diachrony of reciprocals.

Tests with L1 and early L2 learners have indicated that gender assignment and agreement is acquired easily in French and Spanish, which is plausibly guided by the formal properties that distinguish gender-based noun classes in these languages (Karmiloff-Smith 1978, Pérez Pereira 1991, Möhring 2001). Case studies of bilingual French-German and Italian-German children (Müller 1987, Cantone 1999) have shown that these children develop each gender system independently; no clear evidence for transfer has been found.
The present study investigates the early emergence of Spanish gender assignment and agreement in the spontaneous speech of bilingual Spanish-German children at the one-word stage and at the beginning of the two-word stage. It will be shown that although the morphosyntactic aspects of gender assignment and agreement are not directly affected, agreement within the DP manifests later in the productions of the bilingual children than in the productions of a monolingual control group. This observation can be related to a delay in certain aspects of prosodic acquisition, whereas the formally less transparent German gender system seems to have no impact on the development of the Spanish agreement system.

In some Norwegian dialects, such as older Oslo dialect, the noun mamma ‘mother’ unexpectedly appears to be masculine. The Nordreisa dialect (Northern Norwegian) goes one step further. The word looks like it is masculine, but only in the definite form. This is an unusual “split” because gender mixture is normally based on number, not definiteness (but we find some few corroborative examples in other Norwegian dialects and different, but converging evidence on the Web). The Nordreisa example of mamma is unusual also because agreement targets are affected differently. The preference is for masculine agreement within the noun phrase, but for feminine agreement outside it. This is, therefore, an intriguing example since it combines a split based on definiteness with different gender require-ments according to different agreement targets. On careful analysis, and given strict adherence to the classical, agreement-based definition of gender, the unusual behavior of mamma turns out to confo...

Περίληψη
Διερευνήσαμε τον ρόλο των μη ερμηνεύσιμων χαρακτηριστικών (Chomsky 1995) στην κατάκτηση των ελληνικών ως Γ2. Τα δεδομένα προήλθαν από προφορικές δοκιμασίες που δόθηκαν σε 65 ενήλικες, κατηγοριοποιημένους σε τρεις ομάδες σύμφωνα με τη μητρική τους γλώσσα: αλβανικά, γεωργιανά ή αγγλικά. Οι συγκεκριμένες Γ1 διαφέρουν μεταξύ τους ως προς την ύπαρξη άρθρων και τη συμφωνία γραμματικού γένους, που αποτελούν τους στόχους της έρευνάς μας. Τα αποτελέσματα έδειξαν μεγαλύτερες δυσκολίες στο οριστικό παρά στο αόριστο άρθρο, δυσκολίες στη γραμματική συμφωνία γένους, επίδραση της Γ1 καθώς και του επιπέδου ελληνομάθειας. Τα ευρήματά μας φαίνεται ότι υποστηρίζουν την υπόθεση της ερμηνευσιμότητας των χαρακτηριστικών (πρβλ. Τhe Interpretability Hypothesis, Tsimpli and Dimitrakopoulou 2007).
Λέξεις - κλειδιά: ερμηνευσιμότητα χαρακτηριστικών, ενήλικες με την Ελληνική ως Γ2, άρθρα, συμφωνία γένους

We contrast elicitation and spontaneous speech data in French-speaking children with developmental language disorder and controls, with a focus on determiner phrase and gender agreement. Eight French-speaking children with developmental language disorder and age-matched or language-matched controls were compared on an elicitation task for complex noun-phrases with one or two adjectives (e.g., la petite maison verte ‘the small green house’) and a spontaneous speech sample of 200 utterances containing determiner phrases. Elicitation and spontaneous speech data revealed different profiles in French children with developmental language disorder compared to controls: elicitation tasks revealed specific difficulties with adjective agreement as well as high levels of global error, while spontaneous speech revealed mostly omission and substitution errors, often on determiners. Ultimately, both approaches to evaluating language abilities are complementary, but elicitation tasks might be the most useful tool for rapid identification of difficulties with determiner phrases and agreement in young French-speaking children.

Este estudio investiga la adquisición del género en español como lengua materna. Nos concentramos en los factores que pueden tener un efecto positivo o negativo para el niño que tiene que
descubrir las regularidades para la asignación del género a un sustantivo dado. Estos factores son (a) la transparencia formal del sistema, (b) la marcadez de los rasgos de género y, para un niño bilingüe, (c) el género del sustantivo equivalente de la otra lengua, en este caso el alemán.

Processing research on Spanish gender agreement has focused on L2 learners’ and—to a lesser extent—heritage speakers’ sensitivity to gender agreement violations. This research has been mostly carried out in the written modality, which places heritage speakers at a disadvantage as they are more frequently exposed to Spanish auditorily. This study contributes to the understanding of the differences between heritage and L2 grammars by examining the processing of gender agreement in the auditory modality and its impact on comprehension. Twenty Spanish heritage speakers and 20 intermediate L2 learners listened to stimuli containing two nouns with gender mismatches in the main clause, and an adjective in the relative clause that only agreed in gender with one of the nouns. We measured noun-adjective agreement accuracy through participants’ responses to an auditory task. Our results show that heritage speakers are more accurate than L2 learners in the auditory processing of gender agreement information for comprehension. Additionally, heritage speakers’ accuracy is modulated by their Spanish language proficiency and age of onset. Participants also exhibit higher accuracies in cases in which the adjective agrees with the first noun. We argue that this is an ambiguity resolution strategy influenced by the experimental task.

Previous research suggests that, in the process of language comprehension, different pieces of information are taken as relevant in different languages depending on the language structure. The aim of our research was to determine whether the familiarity of the referential gender influences the process of language comprehension in speakers of the Serbian language. Previous research carried out in the Serbian language showed that the familiarity of the referential gender facilitates the reaction times of the respondents to isolated words (Radanović J., Milin P. 2011. Morpho-semantic properties of Serbian nouns: Animation and gender pairs. Psychology, 44 (4), 343-366). Since the process of language understanding does not take place at the level of isolated words, but at the level of statements encoded by more complex syntactic structures (at least sentences), we aimed to check this in a more natural context.
In Serbian, there is a special class of nouns suitable for exploring this phenomenon. On the basis of their properties (nouns that end with -ica like izdajica in certain situations behave not only as epicene (general gender nouns), but also as hybrid nouns), we formed two experimental situations where, based on (1) the distribution of the targets and controllers of congruence and (2) the type of grammatical agreement by gender, one situation could be specified as the one in which the referential gender was determined, and the other one as the one in which the referential gender was not specified. In addition to theoretical assertions, the use of agreement by feminine (grammatical) gender (the one in which the referential gender is not specified) was determined by a questionnaire (N = 30) in the pilot study.
We conducted a Self-Paced Reading Task on monolingual (from the point of view of language acquisition) speakers of the Serbian language (N = 60), whose results (F1 (1, 118) = 0.009, p = .92, F2 (1, 32) = .02, p = .89) showed that the cognitive system of the speakers of the Serbian language is not sensitive to the differences between the two experimental situations. In order to check this, the experimental situations were compared in a study with a stricter control over new subjects (N = 45), using a more sensitive statistical analysis of linear mixed effects (β = -4.832e-02; SEβ = 3.835e-02; t = -1.26; Pr(>|t|) = .23).
Since the results of both experiments coincided, we can assert with a greater certainty that referential gender is not a significant predictor in the process of language comprehension in Serbian. This could be explained by the rich inflectional morphology of the Serbian language, which enables language information comprehension at the formal level, so the recipient in certain situations does not need to seek additional semantic information.

This research investigates the influence of L1 properties and proficiency level on the acquisition of the Spanish gender agreement system. French and English-speaking learners of Spanish participated in the study. Subjects were divided into four different groups considering their L1 (French and English) and their proficiency level (intermediate and advanced). Subjects completed three different tasks: an untimed grammaticality judgment (UGJT) to measure learners' explicit knowledge, an elicited oral imitation (EOI) and an eye-tracking to assess their implicit knowledge of the Spanish gender agreement system. From this multi-tiered methodology, this research project aimed to examine whether L1 properties and proficiency level influence learners' explicit and implicit knowledge of the Spanish gender agreement. The results from the UGJT suggest that both French and English learners can notice noun-adjective discord. As for the EOI and eye-tracking tasks, only the French advanced learners clearly demonstrated integrated implicit knowledge of gender agreement. Therefore, based on these results, we can imply that implicit knowledge of gender agreement is acquired later and that L1 properties influence this whole process, even at an advanced proficiency level.