Cartoon displayed with special permission from Glasbergen.com
If there were ever to be an SLA ‘Hall of Fame’, prime candidates would be Alberto and Wes, both of them icons of fossilization, each in their different ways.
Alberto (Schumann 1978) was a Costa Rican immigrant to the US, who had little or no contact with English speakers, and whose rudimentary English not only remained static over the 10 months he was studied, but seemed impervious to any instructional interventions. His researcher attributed this stasis to his unwillingness to ‘acculturate’, i.e. to integrate into the dominant English-speaking culture.
Wes (Schmidt 1983), a Japanese immigrant to Hawaii, likewise showed little or no grammatical development over the 3-year period in which his English was tracked, but he did exhibit a high level of communicative competence, including the capacity to manage conversations. He did this through a variety of means, such as paraphrase, use of formulaic language, and repair strategies, impelled always by his conviction that it was as much the responsibility of his interlocutors to understand him as it was his responsibility to make himself understood to them. Given that he was fully immersed in, and positively inclined towards, the host English-speaking community, the apparent fossilization of his English is attributed less to social than to psycholinguistic factors: his inability, for example, to attend to linguistic form.
A third – less well-known – candidate for the Hall of Fame might be Patty (Lardiere 2000): an Indonesian of Chinese origin who lived in Hong Kong before emigrating to the US, marrying first to a Vietnamese and then to an English speaker. Her English was tracked over 10 years, and, while it showed a high degree of syntactical accuracy (i.e. word order), there were no gains in terms of morphological features (word endings, plurals etc). The researcher attributed this to the relatively late age of contact with English, and the concomitant inability to perceive (and hence generalize from) the details. (In language acquisition, the devil is indeed in the details). Patty’s written English, it should be noted, was more accurate than her spoken.
These three stories (and there are many more, less well documented ones) tend to underscore the fact that, as Ellis (2008) puts it: ‘Fossilization is not an all-or-nothing phenomena. First, there is considerable variation in the extent to which individual learners fossilize.… [Moreover] it is perfectly possible for a learner to be fossilized in some aspects of the L2 but to continue to develop in others’ (Ellis 2008, p. 28).
However, of the three case studies cited, only Patty’s really qualifies as a case of true fossilization (and, to be fair, neither Schumann nor Schmidt use the term in their own studies) since hers was the only study that was conducted over long enough a period to suggest that the non-target-like nature of at least some of her interlanguage grammar was immutable. That, after all, is what fossilization means, surely.
Or does it? As Long (2003) reminds us, ‘For many, “fossilization” has simply become a general, non-technical name for non-target-like ultimate attainment, that is, a performance descriptor, a broad-brush method of characterising what a learner did not do, not a competence issue, a matter of what he or she could not do, which is what made the original claim interesting’ (p.513, emphasis added). And he adds, ‘If fossilization is to have value as a construct in SLA theory, it must refer to something other than this general age-related decline in the capacity to acquire any language’ (p.519).
In other words, Long is asking us to distinguish between ‘failure to acquire’ (a widespread phenomenon) and ‘loss of ability to acquire’, a much more slippery notion, since what exactly is it that is lost, and how can you prove a negative anyway? How do we know that Patty won’t suddenly (or even slowly) make a tiny incremental change in the direction of the target? For all these reasons, SLA scholars prefer the less terminal term stabilization: ‘Stabilization refers to a state of L2 development where fluctuation has temporarily ceased. Many L2 learners are familiar with the situation where they appear to plateau, failing to develop despite their continuing efforts to do so, and then make a ‘breakthrough’ sometime later’ (Ellis, op. cit: 30).
Ellis adds that ‘there is also another reason for preferring “stabilization” to “fossilization”. Talk of fossilization positions L2 learners as failures but, in fact, many achieve very considerable success in acquiring an L2’ (op. cit: 13). Indeed, the whole notion of a finite ‘end state’ in language acquisition has been challenged in recent years, along with the idea that native-speaker proficiency should be the measure by which learners should be judged.
From a dynamic systems perspective, that is, one in which language acquisition is viewed as an open and continually evolving system, Larsen-Freeman and Cameron (2008) argue that there can be no ‘end state’ as such, and hence no such thing as fossilization. This is not to say that there is no inertia in dynamic systems, only that it is not permanent. ‘Any fossilized form should be seen against a backdrop of the boundlessness of potentiality that is the hallmark of an open, dynamic system, one in which learners actively transform their linguistic world, not just conform to it’ (p. 142, emphasis added).
In the end, as Ortega (2009: 135) summarises it, ‘the notion of fossilization, while strongly intuitive, has proved to be extremely problematic to pin down’. If I’d thought twice, I might have called this blog ‘The (De-) Stabilization Diaries’ – but, then, would anyone have known what I was talking about?
Be that as it may, and inspired by ‘the boundlessness of potentiality’, I have enrolled in two weeks of intensive Spanish classes. More on that shortly.
References:
Ellis, R. (2008) The Study of Second Language Acquisition [2nd edition]. Oxford: Oxford University Press.
Lardiere, D. (2000) ‘Case and tense in the “fossilized” steady-state’, Second Language Research, 14.
Larsen-Freeman, D. & Cameron, L. (2008) Complex Systems and Applied Linguistics, Oxford: Oxford University Press.
Long, M. (2003) ‘Stabilization and fossilization in second language interlanguage development’, in Doughty, C., and Long, M. (eds.) The Handbook of Second Language Acquisition, Oxford: Blackwell.
Ortega, L. (2009) Understanding Second Language Acquisition, London: Hodder Education.
Schmidt, R. (1983) ‘Interaction, acculturation and the acquisition of communicative competence,’ in Wolfson, N., & Judd, E. (eds.) Sociolinguistics and Second Language Acquisition, Rowley, MA: Newbury House.
Schumann, J. (1978) The Pidginization Process: A Model for Second Language Acquisition, Rowley, MA: Newbury House.
The (De-)Fossilization Diaries 
September 1st, 2013 at 8:17 am
Dear Scott,
This does make a very interesting read. I quite like the idea that my students who may think ( and me as well) that their language development has fossilized can actually at some stage make further progress. It is comforting to think that they have merely stabilized at a particular stage. But what is the trigger that would make them start again?
Following you faithfully:-)
Susmita
September 1st, 2013 at 10:47 am
What is the trigger?
(To misquote Philip Larkin!)
I hope to offer some possible answers over the course of this ‘experiment’.
September 1st, 2013 at 8:39 am
If fossilization refers to an ‘end-state’, could we say of learners who have acquired certain language forms, that the correct forms have become fossilized?
September 1st, 2013 at 10:42 am
Good question, Zee, and one that Long (2003) cites as an example of how problematic the term ‘fossilization’ is: ‘Is fossilization a matter of deviance only, or, as might reasonably be supposed, of correct, nativelike rules and forms, too?’ He cites Elka Todeva, who argues that very advanced learners often overuse certain (correct) forms ‘in contexts where native speakers of the target language use different ones’: arguably, this is a form of fossilization.
September 1st, 2013 at 1:27 pm
Within the different ‘experiments’ you’ll be doing in order to discover what works best in (a possible) ‘de-fossilization’, may I suggest writing this blog in both English and Spanish. I found that when I was pushed out my comfort zone (i.e. day-to-day informal oral transactions and conversations) and forced to write in Spanish (mostly emails at varying degrees of formality), my Spanish took a giant leap. Trying to communicate something in Spanish as effectively as I would in English was a huge challenge and really made me reflect on the language – especially the little details where the devil lies, as you put it. Good luck!!
September 1st, 2013 at 7:29 pm
Eoin, it did occur to me that I should be writing this in Spanish – or at least in English AND Spanish.In the end I decided this might be pushing the boundaries of my comfort zone TOO far! But I do intend to embark on some kind of writing task soon, because I suspect you are right: committing oneself to paper and for an audience really forces attention back on the details, and may be a significant ‘de-blocking’ agent. Whether it impacts on spoken accuracy is another matter – it is interesting (though maybe not surprising) that Patty (the subject of the Lardiere study) was more accurate in her written than in her spoken English.
September 1st, 2013 at 1:36 pm
I had an adult student a while ago who told me that her English had ‘fossilized’ and that there was no point in her studying any more because she felt that she was beyond help. At the time, I had only briefly come across the term and didn’t fully understand it. However, my gut feeling was that can’t be true. I spoke to her about the plateau effect, as you have mentioned, and tried to give her some advice about getting over the plateau. But it was interesting to hear that she thought she was beyond help.
I look forward to hearing what strategies you come up with to help learners realize their potential. and that in fact it is better to think about it as stabilization.
September 1st, 2013 at 7:32 pm
I wonder if ‘fossilization’ is used by some learners (who have heard the term or understand the concept) as an excuse? At least it is one way of attributing failure to biological causes, rather than simply lack of effort.
September 1st, 2013 at 2:29 pm
Thank God for ‘the boundlessness of potentiality’ !
September 1st, 2013 at 2:50 pm
Very interesting, Scott. Thank you! I didn’t know that the notion of fossilization had come under so much scrutiny and criticism.
From a teacher’s / teacher educator’s standpoint, it’s comforting to be reminded / reassured that the process is inherently open and dynamic. It lends a certain degree of SLA support to the intrinsic notion that a good teacher never gives up on his/her learners, no matter how slowly they appear to be making headway (oops…).
But from an adult leaner’s perspective, the far more optimistic notion that (s)he will (almost invariably) go through “a state of L2 development where fluctuation has temporarily ceased” can be a bit unsettling. Today, if I were learning a foreign language for professional purposes (which I suspect applies to 85% of our A1 – B2 students in Brazil anyway), I’d have to ask myself: How long is temporary? Six months? Six years? Is there anything I can do / the teacher can do to speed up the process?
I guess the point that I’m (sort of…) making is that to all intents and purposes, from a learner’s perspective, a temporarily long plateau might, in practical terms, be tantamount to eternal fossilization.
And here’s where your initiative comes in, I think. I’m sure your insights will give us some of the tools that we need to re-interpret the construct of stabilization from the learner’s perspective too and think of ways in which we can address the ever so present and ever so palpable frustration that’s been causing adult students to drop out of their courses at alarming rates – countrywide.
September 1st, 2013 at 7:41 pm
Thanks Luis – yes, the distinction between fossilization and stabilization may be hair-splitting from the point of view of the student who has limited time, contact and resources.
At the same time, how can we give learners a realistic appraisal of what they will likely achieve while at the same time not dashing their hopes from the outset?
Apropos, I’m starting to re-think the relationship between (language) teaching and therapy. The subject of a forthcoming post perhaps?
September 19th, 2013 at 5:46 am
Hi Luiz,
The term stability and “emergent stabilities” is really from complexity theory, which in our field is firmly based on usage based language theory, whereas the concept of fossilization has always been SLA, psycholinguistic and generative based. Hence I would say any concepts you are taking from complexity theory (like emergent stability) are not SLA and these two fields (usage based theory and SLA) cannot be harmonized together.
SLA is part and parcel of the recent history of professional teacher training discourse and research, but it seems to me to be a highly questionable approach to language learning.
Regarding end points and plateaus, nothing lasts forever or is fixed I’m afraid. Most learners in anything will only acquire a limited knowledge or experience of their subject. That is reality, and although it may sound negative it really isn’t.
September 19th, 2013 at 5:58 am
As a final note, SLA because of its generative background takes an inherently negative view of an L2 learner. An L2 learner will never be the idealized native speaker, hence concepts like “fossilization”.
September 1st, 2013 at 3:54 pm
This is such an intriguing topic. It almost feels like Professor Langdon’s uncovering the mystery behind the Holy Grail 🙂
I have always kind of felt that the best way to avoid fossilization is to constantly combine learning and acquiring. I believe that if a learner consciously learns a language but has no contact with it outside the class and has no real purpose to use it, the process will slow down sooner or later (by the way, I love Eoin’s idea that you could write this blog in Spanish). And vice versa, if I only pick up the language without any conscious effort to learn it, the results will only be satisfactory.
To me the whole concept of fossilization appears rather subjective, though; I feel that my English is fossilized because I would never be able to write a good novel in English, even though I’ve spent all my life learning it. But someone’s fossilized target language is probably very different from my fossilized English. Isn’t it natural that people always want more than they’ve got; that they want to cross that imaginary line and go further? There’s always one more step they can take to improve. And if somebody is quite happy and doesn’t want to take that step, can we call their language fossilized at all?
To conclude this, I would be an optimist: there’s always a new word, collocation or an idiom I can learn, even though I may feel that the language I’m learning is fossilized.
I’m looking forward to hearing about your improvements in Spanish.
September 1st, 2013 at 7:51 pm
Thanks, Tichahana, for those thoughtful comments. The point about there being no finite end-state – and that it is a human trait to feel that there is always ‘room for improvement’ – is interesting. I wonder if the world is divided into the optimists – those who want to rise to the challenge of going that extra mile – and the pessimists: those who are demotivated by the realisation that perfection is elusive. If you fall into the latter group (like I do – or did) then you also have two choices: to be content (as you say) with your lot, or to always feel like an underachiever. I have to say that for a long time I think I was in the latter category, but didn’t quite see any way out.
September 1st, 2013 at 6:29 pm
Hi Scott,
Your point that language learning is dynamic and doesn’t have an end is really important. When I was doing my master’s (quite a few years ago) I used to get frustrated when reading about studies that described how long it took for certain subjects to achieve competence in a language. I couldn’t understand what they meant by achieving competence. Even native speakers (like me) continue to develop their language skills throughout their lives. So I’m pleased to read that Larsen-Freeman and Cameron have introduced the dynamic concept to applied linguistics.
As for fossilization, does it just mean “stopping getting better”? I’ve tended to use it to describe learners who have learnt language incorrectly and then find themselves unable to unlearn these items. But an inability to unlearn is very different from an ability to learn.
I much prefer the term stabilization. Can we all start using that instead please?
Good luck with your intensive course. Have you been level-tested yet?
September 1st, 2013 at 7:55 pm
Hi Steve: your wording (‘an inability to unlearn is very different from an ability to learn’) captures much more succinctly than Michael Long does the issue at the heart of the fossilization debate. Technically, fossilization describes ‘an inability to unlearn’, but is used loosely to simply mean ‘an inability to learn’ (or, perhaps more accurately, ‘an inability to keep learning’.
I have been level-tested, yes, and I will be writing about that, among other things, next week.
September 2nd, 2013 at 6:12 am
Fantastic- the blog, the experiment, the concept!
As I meander through the paths of traditional approaches, I am unsettled by my emerging views that input is central and output is -comparatively- peripheral. What this means in basic terms is that perhaps you should be hammering the reading and not feel too bad about the lack of writing.
I look forward to your musings on the connections between language learning and therapy. My own views are that they share a lot of similarities. Not least because it seems to me that language teachers should be well grounded in psychology in order to do their work well. It strikes me that they should also be even better grounded in neuroscience.
September 2nd, 2013 at 7:21 pm
Thank you for the comment, Your Secrecy. Regarding the question of whether input or output is the more crucial, all I can say is that, in the nearly 30 years I have lived here, I have had plenty of the former (though not of a quality let alone quantity that might have been optimal, admittedly), but am reticent when it comes to output, and thereby (I believe) lies the problem. But I will explore this in more detail once I get an opportunity to manipulate both variables in different settings (classroom, one-to-one, extensive reading etc).
September 2nd, 2013 at 11:17 am
Hi Scott
I’ll be following your journey along the road towards defossilization or destabilization (that doesn’t sound very desirable, does it?) with more than a passing or even professional interest. Your feelings about your (in)competence in Spanish after living in Spain for so many years resonate very strongly with my feelings about Japanese after having lived here even longer. What doesn’t help is that I have acquired a set of survival phrases, which, combined with my nearly native-sounding pronunciation, gives the impression that my Japanese is much better than it really is. At times I try to resign myself by thinking that the “learning window” had shut because I arrived here well after my teens, but I know that I’m fooling myself because the profound sense of frustration and the frequently surfacing sense of embarrassment tell me differently. I have made several attempts at improvement, both structured and unstructured, but the effort required has always been offset by the knowledge that I can get by. At the end of the day, though, or nearly my life, I guess, I know that I desperately want to feel more comfortable in my Japanese language skin. I will therefore be following your journey with great interest and lots of empathy, hope to learn from your experience and perhaps even share mine from time to time. Bon voyage or rather buen viaje!
September 2nd, 2013 at 7:28 pm
Thanks, Yankun … that’s a very interesting point, re the counterproductive effect of having a number of ‘readymade’ expressions to hand, giving a misleading impression of proficiency. I remember when I lived in Egypt, with little or no Arabic, being able to sustain conversations with taxidrivers simply by interpolating ‘backchannel devices’ (wow, really? you don’t say… etc) into their virtual monologues, such that some would even say: ‘Your Arabic is very good!’.
Schmidt (1983) makes a similar point about Wes, i.e. that his control of a large-ish phraseology gave the illusion of fluency, while also helping open up and exploit conversational opportunities. And as long ago as 1976, Wong Fillmore noted that the strategy ‘Get some expressions you understand, and start talking’ gave immigrant children some communicative leverage in the schoolyard. But, as you suggest, if it goes no further than that, it may lead to an over-reliance on these formula, leading in turn to a degree of fossilization, or of stasis, at least. Effectively, this is what happened to Wes – his communicative fluency was achieved at the expense of gains in either complexity or accuracy.
I’ll come back to this theme, because I sense it is key. Thanks for raising it.
September 4th, 2013 at 12:50 am
Excellent, Scott! I like your 3 examples from the literature. Poor Alberto! I think the Acculturalisation Hypothesis which Schumann develops on the basis of Alberto’s lack of success is full of holes, and unlikely to apply to you, anyway. Schmidt’s account is more telling. And very good quick summary of Long. I think Long gives a good argument for preferring stabilisation as a construct; his article in the Handbook of SLA, which you cite, is a must read. Ortega’s VERY good on all this too, as usual following Mike’s line.
Can’t wait for first report!
September 4th, 2013 at 8:03 am
Thanks, Geoff. A last word on Wes: in a recent paper, Nick Ellis mentions that, in a personal communication a few years ago, Dick Schmidt reports that Wes’s English seems not to have changed in the 20 years since he did the original study. This would seem to suggest something more than stabilization!
September 4th, 2013 at 1:23 pm
I imagine Mike Long would say that such “evidence” doesn’t do much to challenge his view that fossilisation as an explanation is “a “black box”, no more revealing than saying that learners cannot progress any further because of “Force X””. He thinks stabilisation is a more relevant object of study because “(i) the existence of stabilisation is not in doubt; (ii) it avoids the methodologically problematic “permanence” issue; (iii) it makes an additional subset of claims empirically testable; and (iv) unless and until solid evidence appears of the psychological reality of fossilization, it lightens the burden of SLA theory and theories by one variably operationalized and as yet empirically unsubstantiated construct”. Still, that was before you started your study. 🙂
November 23rd, 2013 at 11:24 am
Very interesting! We’re discussing this in our Distance Delta Module One course now (under the thread title “Our Lord Scott Thornbury”, which I’m not at all sure the course coordinator realized was meant to be tongue-in-cheek, as I’m a firm ELT atheist).
My main thought on this is basically, it sure seems to me that it’s possible to predict (whether by intuition or science) whether a student is merely plateauing/stabilizing and may develop further at some future point, or whether they’re not likely to develop in this one area in the near future, if at all.
I have some students who have gone through many stabilization periods in their overall acquisition and the “fossilized” errors I’ve noted have survived, set in stone, through all of those periods. My main example is one student of mine who is at a very high level – able to work with high-level authentic materials without having to refer to his native Chinese at all – and yet who can’t seem to remember the 3rd person -s or plural -s (but does remember the possessive -s, interestingly). I mean this is a guy who regularly watches challenging TED talks and reads novels that would be challenging for non-native speakers (The Great Gatsby, for example) and in-depth articles on world politics without trouble, and yet he’ll say things like “My sister and I all went out to dinner” (he has two sisters).
To me, that’s not “stabilizing”, that’s an error that’s survived many cycles of stabilization and improvement, stabilization and improvement, and has remained. Although he’s improved somewhat since I’ve been working with him, it’s difficult to come up with activities that don’t feel infantilizing beyond helping him notice and correct this error.
And it seems this is something I can predict: that his previous use of the language has provided a basis by which I can guess fairly accurately that without strong intervention (or perhaps despite it), the mistake is not likely to go away. Or perhaps it will someday, with help and persistent noticing, but I can predict that it won’t be easy – and as long as his meaning is clear I’m not sure it’s entirely necessary.
Seems to me an error we can get a feel for in terms of it not being likely to go away on its own, and what it might take to help the student develop beyond it, knowing the instruction might still not be successful, is an error that we might still term “fossilized”, perhaps?
November 23rd, 2013 at 12:03 pm
Oh god, I can’t edit that to make it shorter and less rambly. I wish I could.
the “tl;dr” of it is, if a learner encountered something, say, plural or 3rd person -s, early in their learning, and by the time you meet them has clearly had many plateaus and breakthroughs, but that early-learned piece of language has still not been properly acquired/assimilated in their fluent speech – then I feel like perhaps it’s acceptable to categorize that piece of language as ‘fossilized’. Perhaps it’ll be fixed later on, perhaps not, but at that moment it is a fossil.
I mean, if several stabilizations and bursts of improvement have happened between when they learned the language item in question, and when you notice it’s still a problem, then it’s fairly easy to predict that future plateaus and breakthroughs are not likely to do much to right that error.
November 24th, 2013 at 10:56 pm
Hi Jenna, thanks for your comment. Certainly it does seem that certain non-target-like forms (let’s not call them errors) do seem to have established themselves in a way that seems impervious to instruction or correction (although does this same student make the same morpheme omissions when he writes??). As I said, in my original post, the case of Patty would seem to bear this out, not least because the study was conducted over so long a period. But, as Long said, we still need to know if these seemingly fossilized forms are evidence of what the learner did not do, or what he/she could not do, before we can characterize them as evidence of the internal grammar having ‘closed down’.
November 26th, 2013 at 2:07 pm
So you don’t think there’s any way to extrapolate from what the learner did not do to what the learner can not (or will not – I think most people “can” do far more than they may realize or demonstrate) do?
We don’t do writing in his class. It’s a highly specialized one-on-one and he only wants to read and discuss interesting authentic materials (newspaper and magazine articles – the longer the better – TED talks, Daily Show clips as long as they’re not too vulgar, that sort of thing). Considering his situation – it’d take too long to explain – I’ve decided this is fine. I’ve seen a few examples of his writing when we go over presentations he has to give (I’ve tried weaning him off writing a script, which has only recently proven successful) – he writes these same ‘non-target-like forms’ intermittently, but not as consistently as when he speaks. When he speaks from these notes, he omits the plural and verb-inflecting -s even if he’s written it.
November 26th, 2013 at 8:33 pm
No, I don’t think you can extrapolate, logically speaking. The fact that I didn’t put the cap back on the toothpaste, even on successive occasions, doesn’t mean that I can’t (or, indeed, that I won’t). Same applies to third person s!
November 27th, 2013 at 10:13 am
Footnote to the above: Selinker (who gave us the term) later distinguished between ‘soft’ and ‘hard’ fossilization, soft F being that which is remedial, and hard F being that which is impossible to eradicate. But the question still remains: ‘impossible’ over how long a period and under what conditions?