Beyond instructed L2 grammar acquisition: Theoretical insights and pedagogical considerations about the role of prior language knowledge

The prior language knowledge of learners for whom the target language is not the first foreign language poses a different starting learning situation that should merit pedagogical attention. The present paper seeks to contribute to the question of which pedagogical considerations can be made in regard to the role of prior language knowledge beyond instructed L2 grammar acquisition. Moreover, it fills a significant gap expanding the limited existing pedagogical options that instructors have at their disposal when it comes to teaching in classrooms where one foreign language is simultaneously chronologically first to some and second to others. Starting with (combinations of) existing theoretical accounts and associated pedagogical aspects (such as explicit information, negative evidence, metalinguistic explanations, grammar consciousness raising, and input enhancement), a recently developed method (Hahn & Angelovska, 2017) is discussed. The method acknowledges equally the three phases of input, practice and output and is applicable in instructed L2 grammar acquisition and beyond.


Introduction
The role of prior language knowledge has been a point of endless discussion in both applied and generative linguistics (Weinreich, 1953;Zobl, 1980) and in the field of second language acquisition, resulting in various theoretical accounts about transfer as a constant characteristic of interlanguage (Selinker, 1972). The prior language knowledge of learners for whom the target language is not the first foreign language poses a different starting learning situation, which should merit pedagogical attention. However, it is still not clear why the pedagogical treatment of this issue has generated relatively little attention among researchers and practitioners.

Review of existing theoretical insights
The recent research results point to general lack of robust evidence (as assessed through performance measures and acceptability judgment tests) for any benefits of prior language learning experience and knowledge. The available research points out that prior language experience and knowledge could be beneficial only at certain stages in the L3 acquisition (henceforth L3A) process (Dewaele, 2010;Peyer, Kaiser & Berthele, 2010). Moreover, the level of additional language proficiency was found to influence the benefits in L3 outcomes resulting in greater benefits (e.g., Cenoz, 2003). In a more recent study testing L3 learners at the early stage, Grey, Williams and Rebuschat (2014) found weak effects for the potential benefits of prior language knowledge in L3A. They concluded that the proficiency level has not been reached so as to contribute to L3A and that the role of prior language knowledge is still fuzzy when it comes to incidental learning of an additional language. They did not include any control group with truly novice learners with no prior language learning experience.
The influence of prior linguistic knowledge on the L3A process has dominated the field recently resulting in explanations based on numerous influential factors such as the L1 factor, the L2 factor, typology, psychotypology, recency of use, proficiency levels in all languages of the multilingual individual's repertoire, the characteristics of the universal language processor, length of exposure to the target language, and order of acquisition, among many (see De Angelis, 2007 for a more detailed review). The existing transfer models are based on explaining which of the prior languages will take over the leading role in L3A. Similarly, some studies concluded that only one language (L1, e.g., the L1 factor; Hermas, 2010Hermas, , 2014aHermas, , 2014b, or the L2, e.g., the L2 status factor; Bardel & Falk, 2007Falk, 2011;Falk & Bardel, 2010) can have a privileged role in L3A. On the other hand, some consider the possibility that both the L1 and the L2 are the source of transfer in L3A (e.g., the cumulative enhancement model [CEM] ;Flynn, Foley, & Vinnitskaya, 2004;Berkes & Flynn, 2012; the typological proximity model [TPM]; Cabrelli Amaro & Rothman, 2015;Rothman, 2010Rothman, , 2011Rothman, , 2013Rothman, , 2015. What these studies have in common is the position that acquiring a third language in adulthood is constrained by universal principles and that the type and point at which transfer will happen is determined not only by linguistic, but also by cognitive factors. The predictions of these models rest around explanations of cognitive and structural linguistic nature. They all mainly account for the mental representation of the prior languages in L3A by using interpretation or production data (either spoken or written data) with a focus on either elementary or intermediate learners. Taking into consideration the relevance of the prior language knowledge in L3A, research focused on pedagogical approaches resulting from its results is surprisingly scant.
Though the term prior language knowledge itself is self-explanatory and recognized as a possible predicting factor for transfer in language acquisition, it seems relevant to acknowledge the importance of its link to linguistic experience. The latter has been defined as "the timing, type, modality, frequency and amount of exposure to relevant input and use of the language" (Montrul, De La Fuente, Davidson, & Foote, 2012, p. 88). All these components mentioned in the definition are important when contemplating practical considerations of the role of prior language knowledge in instructed L2 grammar acquisition and beyond. Related questions of practical importance are for example the following: When do we focus on tackling what type of prior language knowledge in instruction? What modality do we choose for its pedagogical treatment? What is the role of frequency in the pedagogical treatment of certain transfer phenomena? What type of input evidence (positive and/or negative) do we need to tackle transfer in instructed grammar acquisition? The answers to these questions are closely related to the amount and quality of prior language use on a daily basis. More precisely, the dominant language used daily, also called "the language of communication," has been defined as "the spoken language used more frequently by the participants at home (with their parents and siblings), at schools (in the classroom and in the playground) and in social contexts (with friends, relatives and other people)" (Fallah, Jabbari, & Fazilatfar, 2016, p. 262). Clearly, there is a need for teachers to explore these realities about which languages are the dominant languages of their learners and to ensure that these realities find their reflection in instruction. In this line of research, it is important to acknowledge that teachers' beliefs about the role of prior language knowledge in language learning (De Angelis, 2012) were given full attention, or as Wright (2002) put it: "A linguistically aware teacher not only understands how language works, but understands the student's struggle with language and is sensitive to errors and other interlanguage features" (p. 115). Given the fact that knowledge of teachers' beliefs is crucial to understanding teachers' actions and choices in the classroom, Haukås (2016) explored Norwegian language teachers' beliefs about multilingualism and the use of a multilingual pedagogical approach in the L3 classroom. The study included data (interviews) collected from 12 teachers of French (N = 4), German (N = 2) and Spanish (N = 6) analyzed by using qualitative content analysis. One of the main themes that emerged from the analysis as most relevant referred to the collaboration across languages. Teachers believed that it could enhance students' language learning; however, no such collaboration was reported. Similarly, the aspect seems even more complex because "a learner's linguistic background matters, much more so in the case of L3 learners (L3ers), who start the acquisition process with a larger amount of linguistic experience and mental representations to draw from" (González Alonso & Rothman, 2017, p. 281). Not surprisingly, the relatively new field of instructed L3A has developed within the recent years as a field on its own covering different lines of research and theories explaining the acquisition of languages beyond the first foreign or second language. However, what remained on the to-do lists of many of us is the need to offer plausible implications for teaching. After having gained some highly interesting results about transfer in L3 acquisition, one very important challenge for researchers is the question of how to apply these highly interesting research results in pedagogy.
In both fields of instructed SLA and L3A 1 teachers are left alone, knowing neither how to predict possible transfer phenomena nor how to deal with negative transfer phenomena. Unarguably, enhancing facilitative transfer is helpful for learners. But as it happens irrespectively of whether one does or does not have the metalinguistic awareness to realize it, no specific typologically tailored interventions are needed. In this line of research, the trend which started a while ago focusing on creating synergies in L3 learning is worth pursuing further (Hufeisen & Neuner, 2003). However, we should not stop here and neglect the problems which follow from negative transfer. Moreover, from a cognitive point of view, it is less problematic for learners and teachers to establish links based on positive transfer between the languages than it is to overcome negative transfer phenomena. Relatively less research exists about how to deal with negative transfer from a pedagogical point of view.

Existing pedagogical considerations
For understandable reasons, a relatively unexplored topic within a relatively new field of research does not offer a lot of possibilities for situating it within an extensive literature review or even building upon existing pedagogical suggestions.
Understanding the reasons for choosing one certain pedagogical suggestion requires an understanding of the distinction between the types of knowledge about/in an L2. Regardless of the L2 acquisition theory, L2 researchers distinguish between a conscious, controlled type of knowledge (i.e., learned and explicit) and intuitive, automatic knowledge (i.e., acquired and implicit) (Ellis, 2005;Wong, Gil, & Marsden, 2014). What is less known, however, is the relationship between these two types of knowledge. Mainly, the existing pedagogical suggestions rest upon such views that either support the no-interface position, the weak-interface position or the strong-interface position. Linking these to the usefulness of negative evidence (i.e., explicit information), the following directions can be stated: Explicit instruction about language cannot affect linguistic competence (no-interface position), both "acquired" and "learned" knowledge (though different) are inherent in L2 grammar (weak-interface position), and explicit instruction about language can affect linguistic competence (strong-interface position). In what follows, pedagogical choices deemed relevant for handling prior language knowledge within both explicit (e.g., metalinguistic knowledge, negative evidence, explicit information, etc.) and implicit grammar instruction (input enhancement) will be reviewed.
Metalinguistic knowledge, when verbalized, rests upon language aspects expressed by using language means, such as metalinguistic explanations, metalinguistic awareness-raising, metalinguistic terminology, and metalinguistic comments. Elaborating on the role of metalinguistic knowledge in grammar acquisition and likewise the role of raising metalinguistic awareness in grammar instruction presupposes a discussion about the benefits of tackling negative evidence in L2 grammar acquisition and instruction and beyond. Negative evidence is the information about what is not possible in a language; clearly, this type of information is explicit information. As expected, we need to discuss its usefulness as accounted for in theory and research. According to Larsen-Freeman (2014), "negative evidence gives students the feedback they need to reject or modify their hypotheses about how the target language is formed or functions" (p. 262). Hence, this implies that negative evidence would be needed for restructuring in learners' minds to happen (White, 1991). Nevertheless, in the case of the study by Schwartz and Gubala-Ryzak (1992), the proof that negative evidence would contribute to restructuring and, in their case, unlearning the verb-movement, was not found. More recent accounts regarding verb movement (more precisely adverb placement) were offered by Rankin (2013), who argues for an insufficient role of positive evidence in the case of English word order as it is "not robust enough to force grammar restructuring" (p. 74). The reasoning he offers for this is attributed to the limited input (both quantitatively and qualitatively) in instructed English as a foreign language context in comparison to immersion contexts. Similarly, Westergaard (2003) investigated the verb-second 2 (V2) in English by Norwegian children and attributed the problems to "the frequency of occurrence of certain structures in English and the controlled input in a pedagogical setting" (p. 84). The insufficiency of the input in such cases leads to a situation where learners even at higher proficiency levels exhibit complex constraints. Thus, for V2 in the case of English acquisition by L1 German learners, Robertson and Sorace (1999) showed that even the grammar of very advanced learners of English can still exhibit the V2 constraint. Obviously, the problem arising from such complex constraints poses further difficulties for its pedagogical treatment, and such complex constraints in grammar "do not lend themselves to an accurate and simple pedagogical treatment that learners can subsequently draw upon" (Rankin, 2013, p. 74). As a result, Rankin (2013) suggests that word order should be explicitly taught, advocating for grammatical enquiry skills through grammaring. Grammaring is a term introduced by Larsen-Freeman (2003) and it refers to students' analysis of the target language structure with the aim to build hypotheses about the form, use and meaning of those structures.
Another pedagogical approach similar to grammaring is grammar consciousness-raising (GCR). It was first introduced by Rutherford and Sharwood Smith (1985) and it refers to increasing learners' awareness through interactive components and opportunities to engage in meaningful interaction and to negotiate meaning with the idea that interaction is essential to language acquisition (e.g., Ellis, 1997;Nunan, 1993). GCR was defined by Ellis (1997) as a pedagogic activity where the learners are provided with L2 data in some form and required to perform some operation on or with it, the purpose of which is to arrive at an explicit understanding of some linguistic property or properties of the target language. (p. 160) The focus in a GCR task is the grammar feature/structure; at the same time, it is also the material about which interaction is developed, so learners are not pushed to produce the target feature/structure but only to exchange information about it, which usually happens through the employment of metalinguistic comments and explanations. The original GCR task was readapted later by Simard and Wong (2004) beyond an L2 situation. More precisely, they required learners to use their prior language knowledge in a pair work activity in which each person was given a set of sentences to read aloud to the other without showing the written sentences. The word order of the sentences could be correct or wrong, and learners had to negotiate and to come to a mutual decision. Fotos (1993) compared the effects on noticing 3 as generated by three different tasks (GCR, teacher-fronted instruction and a communicative task) and reported better effects for noticing over time for the GCR group.
A more recent study, related to the concept of GCR, was conducted by Angelovska and Hahn (2014). They provide implications for the teaching and learning of L3 grammar based on analyses of language reflection sessions between a foreign language teacher and an L3 learner of English. The data they use is based on two sources: learners' written assignments, corrected by native speakers of English, and the corresponding language reflection sessions with the language teacher. The participants in the language reflection sessions have no professional linguistic background, are of different mother tongues (L1s) and of different levels of L3 proficiency, and have the same acquisition sequence of L2 (German) and L3 (English) languages. They have learned English as part of a flipped classroom course at the Language Centre of the University of Munich, Germany. This course includes home self-study, face-to-face language reflection sessions with a language teacher, face-to-face group communication sessions and an online participation in the Moodle learning platform. The aim of the "language reflection" session is to make learners aware of language problems they encountered during the L3 writing process and foster grammar learning by raising their language awareness through exploration (cf. Angelovska & Hahn, 2014). Raising learners' language awareness is a great challenge for the L3 language teacher. Angelovska and Hahn (2014) analyzed the performance of the language teachers by identifying concrete instances when opportunities to raise language awareness were met and/or missed and looked at L3 learners' ways of reflecting on grammar. Some of the questions they addressed are who of the L3 learners "notices" what kind of gaps in L3 grammar knowledge they have; and if yes, how do they do this. Their central analysis was based on face-to-face reflection sessions in which one language teacher focuses the whole session on providing opportunities for language awareness-raising for one L3 language learner. There is still a lack of a framework entailing plausible suggestions of how to handle the classroom situations with many L3 learners who do not necessarily share one and the same language repertoire.
As mentioned, the prior language knowledge in the L3A process seems to pose differing starting conditions for this L3A process. Sanz (2000) compared monolinguals (Spanish) and bilinguals (Spanish/Catalan) acquiring English and found that bilinguals' heightened metalinguistic awareness is a result of the exposure to literacy in their two other languages, which was explained as focusing on form and attending to the relevant features in the input. When investigating the relationship between factors of biliteracy and L3 proficiency, Rausch, Naumann, and Jude (2011) showed that bilingual biliterate Turkish-German secondary school students achieved better results than German secondary school students on L3 reading proficiency and metalinguistic awareness. Clearly, metalinguistic explanations and metalinguistic awareness are considered important aspects in all three pedagogical approaches of explicit instruction, grammaring and GCR. However, it is another question whether metalinguistic explanations (i.e., explicit information) contribute to acquisition. This question takes us back again to the interface debate. From a generative perspective, as articulated by Schwartz (1993) and more recent accounts (González Alonso & Rothman, 2017), learned knowledge (metalinguistic information) cannot serve as input within the grammar acquisition process and cannot contribute to the development of acquired knowledge.  further clarify this by explaining that although exploration based on metalinguistic information such as this sentence [in the context] is ungrammatical may result in learning a concrete fact about the functioning of the L2, the hypothesised language module cannot use the content of this fact as a trigger for change to happen to the underlying grammar. (p. 204) Clearly, obtaining more reliable results about the type of training (explicit versus implicit) would be needed to account for the type of knowledge that is being processed and acquired in the acquisition process. There is a large database of studies in the field of second language acquisition primarily making use of only offline measures to investigate how learners comprehend and process input. However, Benati (2016) calls for the use of online measures (within the processing instruction framework) to obtain more direct results about the acquisition of implicit knowledge with the argument that while behavioral studies can track down only the automatization of second language knowledge, online measures (eye tracking, EEG and fNIRS) would track the online processing, internalization and proceduralization of this knowledge. Similarly, Morgan-Short, Finger, and Ullman (2012) in an event-related potentials (ERPs) study examined the effects of no exposure to a target language and compared two different groups (explicitly and implicitly trained). On the one hand, their results showed that the explicitly and implicitly trained groups did not differ in their performances. 4 In terms of electrophysiological signature, the implicitly trained group showed more native-speaker's performance. These findings clearly justify Whong et al.'s (2014) call for measures of implicit knowledge. Hence, without doubt all researchers and practitioners are interested in fostering implicit rather than explicit knowledge. It seems reasonable that drawing learners' attention to relevant input elements must happen within an implicit intervention, 5 technique or method. One such technique is input enhancement.
Input enhancement (Sharwood Smith, 1991) refers to making certain input elements salient so that L2 learners can notice a feature. For that purpose, the input features entail typographical cues so that the target form is enhanced and its visual appearance in the text is altered (italicized, bolded, underlined). The aim of the typographical change of the target form is to "enhance its perceptual salience" (Wong, 2005, p. 49). In the meta-analysis of the effects of textual enhancement, Lee and Huang (2008) concluded that enhancement is effective and better than no enhancement at all. It seems important here to stress that although the aim of input enhancement techniques is to make learners notice the target features, it does not necessarily require the learner to connect form and meaning or function (cf. VanPatten, 2012). Based on the above discussion of the role of explicit information, it seems that even if learners were explicitly told to pay attention to the enhanced part, it would not make any difference (as found by Shook, 1994); however, this needs further investigation. Even things such as the level of input (sentence or discourse) may impact the effects of textual enhancement. For example, Wong (2003) found out that learners direct their attentional resources to the enhanced elements when the task entails sentence-level input, a condition in which comprehension is less demanding than it is the case with discourse level. What still seems unobserved is the role of the prior language knowledge in relation to decisions as to what needs to be enhanced in the input provided-a pedagogical question relevant for practitioners.
Taking into consideration findings from the accounts discussed in this section, it seems important to convert (aspects of) these theoretical accounts into concrete practical suggestions of pedagogical importance.

Opening new windows: The input-practice-output method (Hahn & Angelovska, 2017)
Incorporating aspects of the above-mentioned theoretical accounts into a concrete method that equally acknowledges the three phases of input, practice and output thereby taking into consideration the associated factors poses a real challenge. Complexities and overlapping interactions of these associated factors need to be considered to further derive meaningful teaching implications. Although we acknowledge the importance of heterogeneity in a language classroom, the starting point of this section is neither to find a solution to the diversity in current language classrooms nor show how to handle it better. We rather present a recently developed method (Hahn & Angelovska, 2017) that approached the topic of negative syntactic transfer and is applicable both in instructed L2 grammar acquisition and in L3A. The starting point in this method is acknowledging the typological differences within one learner's linguistic repertoire and the emphasis on raising teachers' awareness of what triggers negative syntactic transfer when English is L3 and German L2 (or German is L1 and English L2), irrespective of learners' individual differences.
Starting with the existing results and non-results on L3A (transfer), their propositions and explanations and based on data from the study by Angelovska (2017), Hahn and Angelovska (2017) concentrated on the importance of raising teachers' linguistic awareness of phonologically and/or semantically similar adverbs in L2 and L3 (i.e., triggers for verb placement transfer) and proposed an integrative method. They integrated existing theoretical accounts and results from an instructed L3A scenario with L1 Russian-L2 German-L3 English (focusing on V2) into a specific teaching framework thereby developing practical applications as part of the input-practice-output (IPO) method. 6 They show that a combination of these three phases allows for an individual, learner-centered and teacher's awareness-based framework for instructed grammar acquisition.

Target feature in the IPO method: The V2 (verb second) phenomenon
Apart from English, all Germanic languages are V2. V2 refers to the phenomenon where the finite verb is required to appear in the second position of a declarative main clause preceded by a single arbitrary constituent. The initial sentence arbitrary constituent is not (necessarily) the subject. For example, German will require the verb element to be the second constituent in the main clause when the sentence initial constituent is an adverb (German with English glosses): (1) Zuerst dachte ich er hat Interesse. at first thought I he is interested Adv. V S O It is obvious that German requires that the subject be preceded by a verb in main declarative clauses if the initial constituent is an adverb (XVS order); thus, the sentence takes the order of adverb-verb-subject (AdvVSO). In contrast, in English a main declarative sentence with a fronted constituent, such as an adverb, will not result in subject-verb inversion, that is, the English sentence will keep its rigid SVO order, as illustrated in (2): (2) Yesterday I kissed him.

Adv. S V O
Unlike German but like English, Russian does not allow either VSO or XVSO orders, though the XVOS order is possible. In Russian, if the subject is preceded by a verb, the subject must appear in sentence final position (see Erechko, 2003). See Examples (3)-(5) for an illustration of possible word orders in Russian in relation to which a constituent is stressed and thus meaning-carrying (i.e., the final constituent in the sentence is always stressed): ( It is obvious from the presented examples that Russian and English share the same word order in declarative main clauses with an adverb as a sentence-initial constituent in contrast to German, where subject-verb inversion is required. These contrasting scenarios pose a challenge in terms to L2 transfer of V2 in L3 topicalised main declarative clauses.

The components of the IPO method
The proposed pedagogical method incorporates the phases of input, practice and output. The input phase is the most important phase of the IPO method. During this phase learners are exposed to meaningful input, which is specifically tailored to their linguistic backgrounds. In addition, the triggers for syntactic transfer, that is, the phonological similarities, are the key component according to which the teacher is guided when creating the input material. Let us look at a transfer example where it is very difficult to distinguish whether it is the semantics or/and the phonetics that triggers negative syntactic transfer: In the morning went (V) she (S) to school.
One could assume that the phonetic similarity of the onset syllables in morning and the German Morgen triggers transfer from German, and thus the learner realizes this transfer on the syntactic level by producing a V2 structure. However, it is hard to distinguish whether the phonetic similarity triggered the transfer on the syntactic level directly or indirectly, that is, via a much more complex process. For example, the phonetic similarity triggers semantic concepts from the L2 German, and then the semantic concepts trigger the syntactic transfer. As it is not always easy to distinguish clearly what the trigger is, the latter solution seems to be quite plausible in cases where there is an overlap of phonological and semantic similarity, or at least it cannot be simply ignored as a possible explanation. Therefore, Hahn and Angelovska (2017) propose enhancement of these features of the input that are relevant to the respective syntactic transfer. In our case, the phonetic similarity of adverbs between the languages in one's language repertoire triggers the wrong choice of verb placement in the L3. Thus, an illustration of a phonetic rule system in which an enhanced input is shown through parallels and differences between L1 and L3, as well as L2 and L3 plus the respective combinations with semantic concepts is needed as a starting point. Taking this into consideration, in the following we offer suggestions for developing materials for both the practice and input phase. First, teachers need to start with finding examples. In our case, situations where phonological resemblance of preceding sentence constituent triggers negative syntactic transfer on the word order level, that is, V2-constraints in L3 English. The next step will be to present an analysis of the structure of the sentences in the two different languages and point out differences or similarities. After having found the appropriate examples, the triggers must be enhanced according to principles in line with the MOGUL (modular on-line growth and use of language) framework (Sharwood Smith, 2009;Truscott & Sharwood Smith, 2004). The MOGUL framework with a weak-interface position implicates that both "acquired" and "learned" knowledge (though different) are inherent in L2 grammar. It shows that linking accounts of online processing with accounts of consciousness shows whether and how making input salient plays a role in triggering language acquisition. In the case of input enhancement (where the perceptual, conceptual and affective modules can be directly affected), strong perceptual representation should have a better chance of triggering phonological and syntactical growth using textual enhancement (boldface, colour), oral enhancement (stress, volume, length, intonation, pauses, gestures, facial expressions) and input flooding, that is, increasing the number of occurrences rather than altering the salience of an individual instance (Sharwood Smith & Truscott, 2014). Besides enhancement in a written text, an oral enhancement for our case can also be used, either in isolation or even in combination with textual enhancement. For example, the teacher can stress the intonation of the subject preceded by an adverbial phrase in an English sentence: At the moment THEY are busy.
This sentence can be accompanied by a German sentence in which the intonation is put on the verb preceded by an adverb, and not on the subject preceded by the verb. For example: (8) Am Morgen SIND sie beschäftigt.
The pedagogical intervention of processing instruction (Benati & Lee, 2015;Lee & VanPatten, 2003) has demonstrated that it is a more effective pedagogical intervention than traditional instruction and output-based instructional treatments when it comes to form-meaning mapping. Thus, such phonological enhancement would seem applicable in this phase of the IPO method. However, due to a lack of concrete suggestions as to what modifications would be needed in a context where the prior language knowledge plays a role and where the L3 and L2 may rely on lexical or conceptual links to access the meaning of words in the trilingual mental lexicon (Benati & Schwieter, 2017), such a pedagogical consideration cannot be discussed at present. The second phase of the IPO method is the practice phase, which has the aim to accelerate the process of turning meaningful input into output, that is, enabling longer-term effects of what is processed (intake). In the MOGUL framework, the metalinguistic knowledge plays a role in the discussion of enhancement of perceptual representations as it forms a part of the knowledge used to enhance the representations (Sharwood Smith & Truscott, 2014, p. 272). An example given in Hahn and Angelovska (2017) is a scrambled sentence constituents interactive activity in which learners are required to put the elements into the right order. Although learners are not producing output directly during this phase, they are practicing, and through the collaborative exchange of metalinguistic information and teacher's feedback they are exposed to "grammatically improved enhanced input" (Sharwood Smith & Truscott, 2014, p. 272).
Finally, in the output phase of the IPO method, learners are actively engaged in producing output.

Implications of the IPO method
The insights from the IPO method should find their voices into the development of language teaching materials. These should entail specific input-based activities and materials which sensitize learners to using the correct word order. For this purpose, precisely these similar adverbs should be pro-actively used in the input phase. Although the focus so far has predominantly been on L1 transfer, it seems relevant that teachers should be made aware of the cases where L2 transfer comes into play and overrides L1 transfer. Furthermore, it is important to acknowledge the planning time teachers need when considering the preparation of the materials for all three IPO phases. For example, the enhancement of the input needs to be based on a previous analysis or categorization of specific cases of negative syntactic transfer. Similarly, teachers should carefully plan the enhanced input flooding, as a component of the input phase. Similarly, various modes of input enhancement (colors, sizes, blurbs, underlining, fonts, intonation, etc.) should be used in the design of the materials.
Seeing teachers only as transmitters of content should not be the mere aim of language instruction. The role of teachers for research becomes even more important when teachers are given the responsibility for trying out their materials and contributing to the research theories by reporting on studies conducted within the action-research framework. One suggestion coming out of this paper could be to try out different combinations of the given activities from the three phases of the IPO method with different amounts of input and to report on the results of the effectiveness of these activities in overriding negative L1 and/or L2 transfer. Such questions as whether the importance of the one or other mode and/or sequence and/or amount is effective can only be given by teachers, who have the opportunity of observing and including different sequences of these elements in their real classrooms. So, they should have their voices heard. Only such reports could help in the refinement of the existing theories and pedagogical models of negative language transfer in instructed L2 grammar acquisition and beyond.

Conclusion
To merit pedagogical attention to the role of the prior language knowledge, which poses complexities within instructed L2 grammar acquisition and beyond, this paper contributed to the existing gap through reviewing available pedagogical options and opening new windows to a relatively unexplored topic. It filled a significant gap expanding the small number of pedagogical options that instructors have at their disposal when it comes to teaching in classrooms where one foreign language is simultaneously chronologically first to some and second to others.
We started with a state-of-the-art review of the existing theoretical accounts and associated pedagogical aspects, such as explicit information, negative evidence, metalinguistic explanations, grammar consciousness raising, and input enhancement. Then, we presented a recently developed method (Hahn & Angelovska, 2017) that equally incorporates three phases (input, practice and output) and illustrated its implementation in instructed L2 grammar acquisition and beyond. Ultimately, we proposed that the implications of the method are of three interrelated natures: for the development of teaching materials, for lesson planning and for classroom research.
The multilingualism situation in today's world encompasses linguistic facets of individual, cultural, sociological, educational, and psychological dimensions. Its dynamics are rooted in its conceptualization and in the complexity and diversity of the factors involved in the acquisition, processing and use of the language repertoire from diverse perspectives. Although the integration of the four skills (speaking, writing, reading and listening) should be aimed at when designing materials and activities for the three IPO phases, teachers need to consider learners' age (Angelovska & Benati, 2013;Benati & Angelovska, 2015;Muñoz, 2011;Muñoz & Singleton, 2011) and its relation to their cognitive abilities and cognitive task demands (Benati & Angelovska, 2015). Sanz, Lin, Lado, Stafford, and Bowden (2016) rightly draw attention to "the importance of interactions between pedagogical tools and individual differences in explaining language development" (p. 1). Similarly, investigating the role of working memory capacity, which might play a role in such contexts where metalinguistic information is limited, seems desirable. Ultimately, it seems most relevant to remember that even within quite a homogenous group of learners "there are intriguing new findings on differential learnability of properties within the same groups of learners" (Slabakova, 2016, p. 7). Thus, it seems relevant that such factors should receive detailed treatment in future pedagogical accounts and resulting classroom research studies. The first attempt has been made: A solution for bridging the existing gap between results from linguistic research and language teaching has been offered.