The ‘ Attractive ’ Progressive – Why use the-ing Form in English as a Lingua Franca ?

In SLA literature, the ‘extended’ use of the progressive in L2 English production has often been considered a ‘problem’. The studies have been based on elicited data and explanations for such use have ranged from first language interference to target language input. In this paper, my aim is to look into the use of the progressive in authentic spoken L2 English from a fresh, more communicative perspective, namely that of English as a lingua franca. The main focus is on finding out whether the use of the progressive is really ‘extended’, and if so, does this cause problems in communication. At the end, an explanation for the characteristic use of the progressive in ELF is attempted. The data for the study comes from an academic ELF speech corpus, ELFA.


Introduction
Second Language Acquisition research (SLA), which focuses on how second language learners acquire a foreign, target language, usually adopts the ethos of perceiving learners' deviant use of the target language as faulty and as a distraction to communication.Explanations for such erroneous use are commonly sought in three areas: interference from learners' mother tongue(s), gaps in the learners' developing knowledge of the target language as a system, or factors having to do with the input that the learners have been subjected to either in the target language environment or in the foreign language classroom.
Unfortunately, such research fails to acknowledge the way L2 speakers actually make use of the language for their own purposes (and in their own right, see e.g.Seidlhofer 2001), and that L2 speakers may actually assign different functions to, for example, grammatical structures of the language compared to native usage.As English is now being spoken by many more L2 speakers than native speakers in the world, and used more between L2 speakers than in native-non-native speaker interactions, change in the research paradigm from deploring "abuse" of the language into seeing how the speakers actually utilize the language in interaction is not only justified but also necessary.By finding out distinctive and salient features of English used as a lingua franca (ELF) the teaching of this world language could be adjusted to reflect the real world use at least in two ways.Even though it is not advisable to start teaching the discovered features per se, still, if we are aware of the features that are common among L2 English speakers from various different language backgrounds and that do not seem to interfere with intelligibility in communication, we could re-allocate the time spent on 'cramming' such features and shift our focus on features which do require honing from the point of view of intelligibility.Secondly, acknowledging such features could inform testing practices (especially in cases where the test is said to measure the examinee's ability to function in international settings in English, cf. for example, TOEIC® 15 ) so that we have empirically-based knowledge of what distracts the communication in English in international settings and what does not, instead of having to rely on native speaker intuition about it.Features that are found to be commonly in operation and understood in international use of English could, then, be accepted as normal use of the language even if they deviated from native-like use.
Although it is sometimes claimed that English as a lingua franca does not exist as a variety -either because its speech community is difficult to define, or because there are not enough linguistic features that are shared by all of its speakers -this should not mean that the features are not worth investigating.Whether ELF can be called a 'variety' or not depends wholly on how one defines a 'variety' and is thus out of the scope of this paper.Suffice it to say that ELF research has not suggested there was one monolithic 'ELF variety' spoken whole world over (see e.g.Jenkins 2004: 65) -particularly since homogeneous well-defined speech communities or varieties of any language are becoming increasingly difficult to find in today's 'post-modern' world -and anyhow, it is too early to say what forms the future development of English will take.
What we can describe, however, are the tendencies found in the use of English by L2 speakers.Apart from possibly being utilized in English teaching and testing, such tendencies could also predict future developments for the language as a whole as well.It is no news that large groups of second language speakers can have an influence on even the first language speakers' way of using the language -as McWhorter claims happened to the English language during the Old English period as large groups of Scandinavians invaded the British isles and adopted but also altered the language of their new home (McWhorter 2002).As non-native speakers are in the majority of all English speakers in today's world, they are potentially also causing language change.Because of this, if not for any other reason, it is worth paying attention to the tendencies in L2 speakers' use of the language.
In this article, I present one grammatical construction that emerges from the ELFA corpus (see below) as a salient feature, and possibly a characteristic feature of lingua franca use of English, namely the 'nonnative-like' use of the progressive.The article is based on on-going research on the topic but because of the intriguing nature of the phenomenon it deserves to be acknowledged at this early stage.My aim is to look into the (syntactic) use of the progressive -ing form in a spoken ELF corpus, and try to offer an explanation for its 'non-native-like' use.

Background: The Problematic (?) Progressive
In SLA research, the correct use of the English progressive has often been mentioned as one of the most difficult things to learn in the English language.For example, in Swan and Smith (2001) -an edited collection of "characteristic difficulties of learners of English" (2001: ix) from different mother tongue backgrounds ranging from European languages to African and Asian languages -the progressive is singled out as problematic for almost all learner groups, usually meaning that its use is extended to contexts where it traditionally 'does not belong' such as stative verbs or habits.
Explanations for the phenomenon abound.Often this 'difficulty' is put down to the differences between English and the speaker's mother tongue (see articles in Swan and Smith 2001).The L2 speakers are said not to know how to use the structure properly either because it is missing from their mother tongue as a grammaticalized construction (for example German or Scandinavian languages) or because the use is different in English and the speaker's L1 (for instance in Spanish the use of the progressive is often optional in contexts where it is obligatory in English).Alternatively, reasons have been looked for in the learners' developing target language systems: Platt et al. (1984: 73) found the "extended use of -ing" also in so-called New Englishes (cf.'Outer Circle varieties', Kachru 1985) and suggested this be due to learners' overextending the rules of appropriate use within the system.Lastly, input or teaching related explanations have also been offered.For example, Platt et al.'s (1984: 73) alternative explanation for the phenomenon is the possible "overteaching" of the -ing form at school, and Römer (2005: 173), on the other hand, suggests that the reason may lie in "inadequate descriptions of language phenomena in teaching materials".
It is true that the progressive in Standard English is rather peculiar compared to many other languages that have it because the progressive in Standard English has acquired a number of other meanings and functions besides the general meaning of 'action in progress' (such as indicating temporary action, near future plans, or greater emotional emphasis).Due to this, Comrie (1976: 38) states that: "in English the Progressive has extended well beyond the original definition of progressivity as the combination of continuous meaning and nonstativity."It could also be true that the L2 speakers have "extended the rules" of the language or that the descriptions of the progressive in teaching materials do not match precisely the real life 'authentic' native speaker use of the progressive (as Römer 2005, in fact, demonstrates).
But are these explanations sufficient?What they seem to do is provide excuses for L2 speakers' 'misbehaviour' and not take into consideration that actually L2 speakers could be using the resources of the language for their own purposes.Looking again at the excuses of misbehaviour given above, we can ask: If something was perceived as particularly 'odd' in a foreign language (as the progressive in English in contrast to other languages), would it not rather be the case that such oddity was avoided or replaced with a simpler construction (in this case the simple form)?Or if it is the case that L2 users 'extend the rules', as it were, why do they do so?And finally: are L2 speakers really just trapped by the teaching they have received or the teaching materials they have used even in their daily spontaneous communicative situations, or could it be that they actually are making use of the resources of the language and being creative in their L2?
What is particularly intriguing in the literature on progressives and L2 speakers is the fact that studies on acquisition of grammatical morphemes of English by L2 speakers in the 1970's and 1980's showed that the progressive -ing was, in fact, the easiest verbal morpheme to be acquired by L2 learners, meaning that it was properly attached to and used with verbs early on (as first discovered by Dulay and Burt 1973).Also, it has been found (see, Giacalone Ramat 1997) that the progressive in English seems to be acquired earlier than that in other languages, which Giacalone Ramat ascribes to it being "attention-catching for its frequency in [native speaker] discourse" in English (p.281).Although it is difficult to determine how much native speaker input each learner has been subjected to and what its consequences are for each learner's use of the progressive, there could be a grain of truth in this explanation that I will revisit in the end.
Thus, in the light of this research, a new perspective on the 'problematic' progressive could be adopted: rather than saying L2 speakers do not know how to use the progressive (and for that reason use it in 'wrong' contexts), we could turn the tables and ponder whether there was actually something about the progressive form that particularly attracts the L2 speakers of English from early on.

Increased Use of the Progressive in Native Speaker English
It is also worth noting that the use of the progressive has been on the rise in native speaker English as well.Scheffer (1975) goes back to Old and Middle English and Smitterberg (2005) to 19 th -century data to show how the progressive has become more common in English over the centuries.But even within a shorter time span, during the last few decades, different corpus-based studies have demonstrated increase in its use (see, e.g.Mair and Hundt 1995).On the whole, progressives are found to be more common in spoken than in written language, and according to Longman Grammar of Spoken and Written English (1999: 462) progressives are even more common in American English conversations than in British English conversations.
What could account for the increased use?Although some intuition based studies have assumed the reason for this to be in the increased use of stative verbs in the progressive for specific meanings (cf.I'm seeing ghosts), Mair and Hundt (1995) do not find empirical support for this in their data.Instead, what many researchers seem to suggest as an explanation for the growth in native speaker use is a stylistic one.For example, Potter (1975: 120) refers to speakers' growing desire to make what they say "more lively and vivid", and Scheffer (1975: 110) speculates that besides the fact that some of the functions of the progressive have developed only fairly late, the increased use may be due to the "latitude to convey subtle shades of meaning" that the progressive provides to the speaker/writer.In the same vein, Mair and Hundt (1995: 118-119) suggest the reason be "a textlinguistic or stylistic one" and that it might be led by the affective-emotional use of the progressive (as in You're always complaining) so that "in cases in which the simple form can be used alongside the progressive, the latter tends to be chosen with increasing frequency".
Such stylistic motivations could also be termed impressive motivations -impressive meaning that the speaker uses language in an innovative or unexpected way in order to be noticed (Haspelmath 1999: 1057 and footnotes 9 and 10).What is at stake, in other words, is the speaker's desire to be socially successful and even admired.According to Haspelmath (1999), this type of language use may generally also give rise to language change.If, then, 'impressive' use of language is behind the increased amount of progressives in native speaker English is the same also true for L2 speakers' use of the construction?Let us take a closer look at the matter.

Data and Methods
One of the main goals of the ELFA (English as a Lingua Franca in Academic settings) project based at the University of Tampere and University of Helsinki (see the project Web page at: http://www.uta.fi/laitokset/kielet/engf/research/elfa/) is to find and describe characteristic tendencies in ELF with the help of a spoken corpus comprising authentic ELF speech in academic settings (see Mauranen 2003;this volume).The ELFA corpus that has been compiled for this purpose consists of academic events such as lectures, seminars, thesis defences, conference discussions, and presentations where English is used as a lingua franca between speakers from diverse linguistic backgrounds (though the majority speaks European languages as their L1).The current size of the corpus is approximately 0.6 million wordshalf of which was fully transcribed and rechecked at the time of the retrievals for this paper.Thus, the present study is based on this 0.3million-word-corpus.As a reference corpus, MICASE (Michigan Corpus of Academic Spoken English, see Simpson et al. 1999) with its approximately 1.7 million words was used for comparisons since it consists of similar kind of spoken academic data from a native speaker context (see: http://micase.umdl.umich.edu/m/micase/).
On the whole, academic speech is particularly suitable for ELF research for various reasons including the fact that L2 speakers in academic settings use English as their working language more or less regularly.But most importantly, the academic context per se puts language and its use into serious test as one of the most challenging and demanding spoken (and written) genres in our culture: The speakers have to be able to argue a point, defend their view, be convincing, and discuss generally abstract and complex topics by means of language -there is often very little help, for example, in the physical context to rely on.Managing this type of talk in a foreign language can justly be considered the highest point on an L2 attainment scale.Thus, if we want to look for, for example, new standards or guidelines for L2 testing (instead of the 'educated native speaker'), L2 academic speech with its features lends itself well to establishing the "ultimate attainment" for non-native speakers.
In the ongoing research, my intention is to look closer into the verbsyntactic features of ELF in the ELFA corpus but here the focus is only on the progressive.As mentioned above, half of the ELFA corpuswhich will, nevertheless, be referred to as 'ELFA' in this paper -and the whole of the MICASE corpus were searched with corpus tools for instances of the progressive forms in all tenses (also in the passive voice) both with full and contracted BE-auxiliaries.The retrievals yielded a considerable number of hits that were not genuine progressive forms, including non-finite clauses such as it is restricted to just governments making the decisions, and phrasal expressions such as i think this is worth considering.Such cases were deleted from the data manually.Further, only constructions that could clearly be classified as progressives were included in the data, thus leaving out appositively used particles (i was sitting here thinking; or there's a statement saying), gerunds (what you will be doing is computing the values), as well as all instances of the verb be going to with future reference as it was considered a marker of the future tense rather than a token of the progressive aspect (see, Smitterberg 2005: 26-37 for discussion of the above-mentioned constructions).Finally, instances of repeated verbs where a speaker duplicates one and the same verb in the utterance were deleted (e.g.what you are controlling you are controlling the pressure; and other income is basically subsidising subsidising the kiosk) so that only the first instance of the repeated verb was counted in.After the general mapping out of the instances, all progressives in ELFA and an equal-size random sample of the progressives in MICASE were taken under closer scrutiny.

Overview of the Results: Use of the Progressive in ELFA and MICASE
The post-processed results of the retrievals yielded 1,247 instances of the progressive for ELFA (with 307,411 words altogether) and 12,990 for MICASE (with 1,707,510 words altogether).Normalizing the frequencies to a text length of 10,000 words reveals that the progressive is used in the current version of ELFA approximately 41 times / 10,000 words and in MICASE 76 times / 10,000 words, which suggests that L2 speakers would actually use the progressive considerably less than the native speakers.This is an intriguing result but perhaps the discrepancy need not be as dramatic as it seems at first.At least two data-related factors can be found that can explain (at least some of) the difference.Firstly, the data consist largely of conversational material and debates, and it seems that the native speakers are in the habit of reflecting on the conversation more frequently with immediate metatextual phrases such as do you see what i'm saying?, so the question i'm asking you is, or that's what i'm talking about than non-native speakers in ELFA.This one use increases the number of progressives for MICASE considerably.
The same function is present in ELFA, too, but to a notably lesser degree.Another data-related explanation for the difference may reside in the events included in the two corpora.MICASE seems to include more recordings of lab sessions and other events in natural sciences where demonstrations and hands-on pair or group work are common, resulting in frequent use of the progressive as the on-going processes or immediate actions are being described or commented on, as in: we're heating it up -okay it's getting warm, or look what we're doing.-we're measuring fish, or so R-N-A polymerase, is doing its thing over here.the ribosome's following right after.okay?Nevertheless, a more detailed look into the transcribed documents is required to see whether there are other, truly function-related reasons for the less frequent use of the progressive by L2 speakers.(Although based on the classification of syntactic functions (see below), this does not seem to be the case.) The twenty most common verbs that appear in the progressive in each corpus are shown in Tables 1 and 2 below.If we look at the distribution of the different verbs that take the progressive in each corpus, the results indicate that the use of the progressive in MICASE is concentrated on fewer verbs than in ELFA.In MICASE only 12 verbs account for 50 % of all the progressives in the corpus whereas in ELFA the distribution is wider (and thus more even), with 16 verbs accounting for half of the uses.Also, if we take a closer look at the individual verbs we can see that even though most of the common progressive verbs in each corpus are the same, there are some striking differences in the frequencies of some of the verbs.For example, saying appears to be more than twice as common in MICASE as in ELFA, and doing is more frequent by a half in MICASE compared to ELFA.This seems to tie in with the explanations for more frequent use of the progressive in MICASE that were given above.What is more, corpus analyses reveal that in MICASE the use of the progressive is more clustered, meaning that there are more fixed phrases in which progressives are used in L1 data than in the L2 data.Again, this seems to indicate that the use of the progressives in ELFA is more widely distributed and that the form is used more freely or in more diverse contexts.For a more detailed study, all the instances of progressives in ELFA (N=1,247) and a randomly selected sample of 1,247 progressives in MICASE were classified in their syntactic context using traditional grammar descriptions as a point of reference (notably Quirk et. al. 1985 andBiber et.al, 1999).The majority of the cases in ELFA (87 %) and MICASE (98 %) fell into the typical categories of use for the progressive described in traditional grammars, such as denotations of processes, repeated or continuous actions, currently on-going or temporal actions, near future plans and so on.Comparing the distributions of tenses in progressives in ELFA and MICASE was considered rather difficult due to the above mentioned fact that MICASE appears to consist of more hands-on, immediate action than ELFA resulting in much more frequent use of the present progressive in MICASE and lowering the proportions of other tenses respectively.However, the rank order of different tenses was the same in both corpora, the present tense being the most common, followed by the past tense, the present perfect and the pluperfect, and lastly the future tense.However, in the use of auxiliaries a notable difference was found: in ELFA up to 77 % of all progressives were preceded by a full auxiliary BE whereas in MICASE this percentage was only 42 %, contracted forms being favoured.Regarding progressives in the passive voice, there were slightly more instances in ELFA (3 %) than in MICASE (2 %).
As noted above, the majority of the uses of the progressive in ELFA were 'native-like' but since L2 speakers are accused of extending the use of the progressive, let us now turn to those 13 % (N = 160) of the ELFA progressives that did not fit into the categories provided by traditional grammars.Three types of non-traditional lexico-grammatic uses were found.The categories partly overlap but here each progressive was classified into a particular group only once.The MICASE data was also checked for similar types of uses.

Stative Verbs
One of the verb groups that can take the progressive in ELFA contrary to the descriptions in standard grammars are the so-called stative verbs.These are verbs that denote perceiving (e.g.see, hear, feel), an intellectual state or a state of emotion (e.g.know, mean, understand, like, hate), or a relation (e.g.belong to, consist of, depend on, own).Of all the progressives in ELFA, these account for 3.6 %.Below (see items 1-6) are examples of such use from the ELFA corpus.A description of the recorded event, the speaker's academic status, and the speaker's mother tongue (in italics) are indicated in parentheses at the end of each example.
(1) i mean er er properties and relation are belonging to the same erm ontological general area or cat-category (Philosophy Seminar; Senior Faculty, Danish) (2) hello my name is <NAME> i am coming from er romania where i am a PhD student (Racism in Finland Panel Discussion; Research Student, Romanian) (3) age is a derived property and er is a property which is depending either on other properties or it is derived and computable (Information Technology Thesis Defence; Senior Faculty, German) (4) how is a cirrhotic person looking like , chirrotic person (Internal Medicine Seminar; Senior Faculty, Finnish) (5) the target of the enlargement is to establish the great europe or some western europe because if we talk about the europe it is including russia ukraine belarus romania and so on (Political Science Seminar Presentation; Undergraduate Student, Chinese) (6) then i made a research and er asked er ten students er er why , are they thinking it's too much work for two credit units (Racism in Finland Panel Discussion; Undergraduate Student, Lithuanian) However, this kind of use is to be detected in the MICASE corpus, too (see examples 7-11 below).Occurrences are fewer, though, than in ELFA, accounting altogether only for 1.4 % of all the progressives in MICASE.In the following examples, all speakers are native speakers of American English.
(7) but, the structure there was so different because, i mean at that time, um, Indians were, owning a bunch of, land i guess (Anthropology of American Cities Office Hours; Undergraduate Student) (8) so, first of all [...] she's agreeing with the statement.and then she says, i'm gonna give examples (History Review Discussion Section; Senior Faculty) (9) these are people who, who maybe for the first time they're hearing that you can ask a question in, such a way where y-your ego is safe and the presenter's ego is saved you know (Women in Science Conference Panel; Senior Graduate Student) (10) that's what it's looking like if you've got eighty-four nineteen involved at all (Immunology Lab Meeting; Graduate Student) (11) and if you have a question or a problem it's so much better to talk to the G-S-I ahead of time and, say i'm not understanding this than to, do it after the test (Academic Advising; Staff)

General Validity or Habitual Activity
Another group of verbs that takes the progressive in ELFA (contrary to the descriptions of Standard English) are verbs that denote general validity or truth, or habitual activity (see examples 12-16).These make up 5.6 % of all the progressives used in ELFA.
(12) communication is su-so all-embracive a concept like air that we are breathing (Information Society Seminar; Senior Faculty, Finnish) (13) there are few things that it's polite if man is doing it for example i don't know er waiting in front of the door and that and letting woman go first (Women's Studies Seminar; Graduate Student, Polish) (14) i believe even that using lan-language technology to some extent automate the process that is finding the contexts of the terms that you're looking for [...] i don't know how it's how you're supposed to do it but you can see that it's it it might be possible (Translation Studies Thesis Defence; Senior Faculty, Swedish) (15) in principle every library is free , you the users are paying nothing for library service (Russian Studies Lecture Discussion; Junior Faculty, Russian) (16) i'm not sure if if radical is the is the is the right word , maybe i don't know er bwomen who are er arguing in every situation er with with all the gender stuff (Women's Studies Seminar; Graduate Student, German) Again, there are some instances of such use in the MICASE, too, but to a far lesser extent.Approximately only 0.5 % of the progressives in MICASE occur in habits or so-called 'general truths'.Again, all the speakers in the examples (17-19) are native speakers of American English.
(17) right.and the weft threads are the ones that are creating these selvages.remember?these self, woven edges?(Art History Office Hours; Junior Faculty) (18) where are metastases actually occurring, they're occurring almost predominately almost entirely in the lung.(Biology of Cancer Lecture; Senior Faculty) (19) it's, basically a chronological story of how ideas have changed over the years, about, what, L-S-D and, the phenylethylamines, are doing, neurobiologically, to mediate hallucinogenic effects (Drugs of Abuse Lecture; Senior Faculty)

Points in (Past) Time
In addition to applying the progressive on states or general truths and habits, ELF speakers occasionally use the progressive also in contexts where they refer to points in time rather than a continuous or repeated action (see examples 20-23).Most often the referred event is a point in the past.This use accounts for 2.7 % of all the progressives used in ELFA.
(20) my topic is finland's role in the united nations , just you might wonder i was changing my subject a bit (Political Science Seminar Presentation; Undergraduate Student, German) (21) that came as a result of the establishment of the university which is the agricultural university's forestry branch that was being put up there in late 70's (History of Science Conference Presentation; Senior Faculty, Swedish) (22) you mentioned the role of civil society and [...] that the civil society is somehow lacking in this process er in this political region building process er i was then just starting to think about whether whether the picture would be actually so that there is a lot of civil , like civil society movements [...] but they just don't fit together with this political top-down type of region building (International Relations Seminar; Research Student, Finnish) (23) i mean there will be a big blow and this comes to the point i've been just mentioning before er it can lead to clash of civilisation (Social Dimension of Globalisation Lecture; Junior Faculty, Somali) In MICASE, on the other hand, this type of use of the progressive is practically non-existent.
In addition to the above mentioned categories, there were also other sporadic uses of the progressive in ELFA, accounting for 1 %, that could not be classified.
Studied in their macro-contexts, the three types of non-traditional uses of the progressive presented above tend to occur in monologues or in monologic sequences more often than in dialogues (59 % of the cases).What is also notable is the fact that even though L2 speakers in the ELFA data are frequently found to self-correct their grammar in their utterances, in case of the non-traditional progressives there is virtually no self-correction (apart from one instance).Also, there are no observable signs of this type of use of the progressive causing misunderstanding or miscommunication in any of the instances in the data.
As regards the mother tongues of the L2 speakers in whose speech these uses are to be detected in the ELFA data, they include the following: Germanic languages (German, Danish, Dutch, Swedish), Romance languages (French, Italian, Portuguese, Romanian), Slavic and Baltic languages (Bulgarian, Polish, Russian, Lithuanian), African languages (Akan, Somali, Swahili) as well as Finnish, Arabic, Chinese and Urdu.

Discussion: Why Use -ing Forms in ELF?
The results presented above give rise to a number of considerations.First of all, the data indicates that the phenomenon at hand is not purely a mother tongue dependent feature (as some of the SLA literature has also come to notice) or general interference from an individual L2 speaker's L1 because similar use crops up in the speech of L2 speakers from many typologically different mother tongues.As noted above, also Platt et al. (1984) found a similar phenomenon in their studies of several Outer Circle Englishes in Asia and Africa.This makes it difficult to believe that the reason behind such use of the progressive resides in mere L1 interference, target language input or teaching related factors, as these are sure to differ in different parts of the world.
What is also worth noting is that the so-called 'extended use' of the progressive is not restricted only to stative verbs but is found in other non-traditional contexts as well in ELF.Moreover, on the whole the use of the progressive in spoken ELF seems to be distributed on a wider range of verbs than in native speaker speech where the use is more clustered and concentrated on certain verbs.That is why it is perhaps not surprising that the wider use of the progressive should reflect itself also on contexts that are considered 'deviant' from standard contexts.This is to say that ELF speakers do extend their use of the progressive in new areas but whether they extend the rules of the progressive aspect as such is another matter.I will return to this shortly.as we could see, some of the 'deviant' uses are also found in the native speaker data (though to a lesser degree), which should be reassuring to language pedagogues as L2 speakers are not found to be, after all, so deviant in their use of the language as some of the SLA literature seems to suggest.It is difficult to say why native speakers should extend their use of the progressives to stative verbs (unless this is actually influence from L2 use of English on L1 use!), but the fact that ELFA data also revealed a use that was not found in the native speaker data (i.e.referring to a point in the past with a progressive) gives rise to the assumption that the motivations for similar tendencies (i.e. the 'extended' use of the progressive) in native and non-native do not necessarily arise from the same source.According to the literature on the increased use of the progressive in native speaker speech, the reason for the increase is assumed to be in the growing use of the progressive in contexts that 'allow' it (thus providing a stylistic variant for the simple form), and not really in progressives spreading to new fields.But for the extended use of the progressive in ELF, the stylistic explanation does not seem readily applicable, as in ELF the construction seems to have spread to totally new areas where it is difficult to see what stylistic gains could be made with the use of the progressive (cf.e.g. in the case of general truths).What, then, could explain the 'attractiveness' of the -ing form in L2 use?Let us go back to the findings again.
As was noted in the results, the vast majority (87 %) of the uses of the progressive by L2 speakers were 'standard-like', which suggests that ELF speakers do know the semantics of the progressive and use it accordingly most of the time.However, also a notable number (13 %) of the uses were obviously 'non-progressive' uses of the progressive (i.e.states, general truths and points in time).But rather than saying that L2 speakers over-generalize or extend the rules of where to apply a progressive (i.e. its semantic field), we could say that perhaps they have just assigned it a totally different extra function.Most of the so-called 'deviant' uses occurred in monologues or monologic sequences (59 %) and the progressives seemed to have been made more salient compared to native language use as the great majority (77 %) in ELF use were accompanied by a full auxiliary instead of a contracted one (favoured in native use).Also passive constructions in the -ing form were somewhat more frequent in the ELF data (despite their rather complex nature) than in native speaker data.What is more, the 'non-native-like' use of the progressive did not cause any kind of misunderstanding in the ELFA corpus, and lastly, the data showed that these kinds of uses were almost never subject to self-correction (in contrast to frequent selfrepairs of other grammatical constructions).This last point seems to indicate, at the very least, that such extended use is not found distractive or anomalous by the speakers themselves, and could, at the other extreme, make one wonder whether such use was actually functionally motivated.
Looking at the data and based on the fact that the -ing morpheme is acquired early on in L2 English, I would suggest that the source of the 'attractiveness' of the progressive resides in the grammatical form itself -that L2 speakers have realized its 'communicative' value in interaction.What I mean by this is that adding the ending -ing and the auxiliary BE to a verb (any verb for that matter) gives the verb more prominence and salience in the speaker's utterance.It makes the verb stand out, so to speak, and draws the interlocutor's attention as a 'heavier' periphrastic structure.The verb, on the other hand, has traditionally been considered the most essential part of a sentence, often carrying the core information in an utterance.Hence, it could be argued that the 'extended' use of the progressive in ELF is due to expressive reasons (rather than impressive) -expressivity meaning that the speaker wants to speak as clearly as possible so as to make him/herself understood by others (cf.Haspelmath 1999Haspelmath : 1057)).Especially in monologic speech where the progressives were used more often, expressivity and clarity are of great importance.
In traditional grammaticalization theories, it has been argued for a long time that when a construction is felt to be too weak or 'faded' in its expressivity, speakers introduce a new and fuller periphrastic construction to replace it in order to enhance intelligibility (and if adopted by a large number of speakers, the introduction may eventually lead to language change, cf.Haspelmath 1999: 1050-1051 for a succinct summary of the history of the argument).There is no reason why a similar strategy could not be used by ELF speakers as well.In ELF communication, maintaining mutual intelligibility becomes a high priority among speakers from very different linguistic and cultural backgrounds, and thus safeguarding clarity by different linguistic means would only seem natural.This is also attested in Mauranen (2006) where she looked into the ELFA data in search for misunderstandings and communication breakdowns but found them to be less frequent than expected because ELF speakers resorted to various proactive strategies in anticipation of communicative difficulties.Using grammatical means for such purposes was not studied but highlighting the verb in an utterance could be one such additional means of making sure the message is received and understood by the interlocutor(s) in ELF.In the case of the progressive, the auxiliary (that was more salient in ELFA than in MICASE) gives the listener a hint of the soon following main verb and the morpheme -ing further marks it off.This is not to say, however, that speakers would actively be aware of utilizing such a strategy -as speakers we rarely are aware of the communicative strategies we use -but even the fact that the -ing form is learned and used early on could indicate that also its communicative value is perceived early on.And why stop using such a salient structure even later on if it helps the speaker to make him/herself understood?Thus, I would not say that the frequency of the progressive in (native speaker) speech is the attention-catching factor that makes L2 speakers use the progressive more widely, as Giacalone Ramat (1997) above suggested, but on the contrary that the 'attention-catchingness' of the form is the factor that makes L2 speakers utilize it frequently.
If this is true, ELF speakers can be seen making use of the resources provided by the language for their own purposes.This, I would suggest, indicates that something very essential about the language has indeed been learned, and not the other way round.

Conclusion
Given its wide distribution among L2 speakers around the world, the 'extended use of the progressive' could well be regarded as a characteristic feature of lingua franca English.It was argued here that in addition to the standard-like uses, it may also have its own peculiar function in ELF based on its 'attention-catching' form.Although the SLA literature sees the extended use of the progressive as a problem, there is no support for that in ELFA.Be the reasons for its use whatever they may, it causes no obvious misunderstandings or communication breakdowns, and it does not seem to distract the speakers.What was put forth here is that, in fact, the reverse might be true so that the progressive is actually used for the very purpose of gaining explicitness and expressivity in L2 communication.Also, since feature is in use in Englishes around the world, it is highly unlikely that a new method of teaching the progressive or more authentic descriptions of it could eradicate the 'erroneous' use -and, indeed, why should it be eradicated?As Jenkins (2000: 160) cleverly observes: "There really is no justification for doggedly persisting in referring to an item as 'an error' if the vast majority of the world's L2 English speakers produce and understand it."The extended use of the progressive is a case in point.Why keep on correcting (up to the point that makes Platt et al. 1984 refer to "over-teaching") a feature that L2 speakers find convenient to use in their communication and that presumably has strong motivation to occur in their speech?Why make 'a problem' out of something that could itself be a means to avoid problems?

Table 1 .
The Rank Order, Number of Occurrences and Percentage of the 20 Most Frequent Verbs Occurring in the Progressive in ELFA * Of all progressive forms found in the data

Table 2 :
The Rank Order, Number of Occurrences and Percentage of the 20 Most Frequent Verbs Occurring in the Progressive in MICASE * Of all progressive forms found in the data