Is it time to SLAy PPP?

A recent discussion on Geoff Jordan’s blog, “Why PPP makes no sense at all,” has gotten me thinking more about the role of SLA research in language instruction. Most teachers, in their quest to be pedagogically principled, have taken a more evidence-based approach to their teaching. SLA research is one of the sciences that should underlie how we go about teaching. One of Jordan’s key arguments is that PPP as most conceive it flies in the face of sound SLA research. PPP is often seen as a rigid, linear methodology that assumes learners will learn what is taught, practiced, and produced. Jordan’s oft-repeated response is:

Students do not learn target forms and structures when and how a teacher decrees that they should, but only when they are developmentally ready to do so.

In quoting Ortega (2009), Jordan continues later:

Instruction cannot affect the route of interlanguage development in any significant way.

Geoff’s first argument should not be doubted by anyone. PPP as most think of it is not evidence based. SLA research makes it clear there is a developmental order of language learning. However, leaving PPP aside for the moment, I’d like to focus on the second of his arguments.

Instruction. A great deal of research in SLA has led to an approach to language research and instruction called “Instructed SLA”. In fact, a new journal has recently been founded, Instructed Second Language Acquisition, and I read one of their first articles – “Situating instructed language acquisition: facts about second language acquisition” (VanPatten, 2017) – with great enthusiasm, thinking, “Yes! Finally, we will get a clear articulation of how to teach according to the most recent SLA research. Woot!” I guess I was a bit naive because it appears “instructed SLA” is an oxymoron. While VanPatten’s article echoes the same valid arguments Jordan writes about (there is an ordered development which is quite internal to the learner) there is a clear, depressing implication to this. VanPatten essentially argues that what we see as effective teaching should simply be “considered ‘trivial’ effects (if any) of instruction”. This is because they do not last nor fundamentally alter language learning. He states:

If the ordered development of acquisition is essentially immutable, what is the goal of formal instruction? What does the researcher believe is affected by a treatment in a given study?

Much of the article is based on taking a more critical approach to research, but these initial statements seem to be suggesting that instruction has little to no role to play in language learning. If instruction has no role in language learning, then what is the point? Instructed SLA seems like a contradiction and any discussions of both PPP and TBLT (Jordan’s preferred methodology) are rendered moot.

Of course, there is the argument that a well-designed needs analysis (a cornerstone of TBLT) can help make instruction target students precisely where they are developmentally ready, and this could be what instructed SLA is getting at. However, it seems unlikely. First, unless you are teaching a group of octuplets, learners’ developmentally readiness will likely be vastly different; it is highly individual. Age, cognitive maturity, initial proficiency, previous exposure, and any other number of individual factors can shift what it means to be “ready”. Furthermore, there is the issue of what the needs analyses themselves are looking for. Can they clearly distinguish between a learners needs vs wants? Needs in the SLA sense seem to be highly personal and would require a battery of psychometric tools and individual analyses to be able to determine any one learner’s developmental readiness.  Any grammar diagnostic can only indicate students are or are not proficient in the particular items being addressed, not whether they are ready or not. Needs analyses seem to measure functional needs, not developmental needs. That is, what students need to do with the language. So how is a negotiated syllabus informed by needs (or wants) analysis any different than a linear grammar syllabus if you cannot target learners at developmentally precise times? Sure, students may want to learn how to make a doctor’s appointment or write an expository essay, but it doesn’t mean they are ready for it as SLA sees it.

As de Graff and Housen (2009) write in “Investigating the Effects and Effectiveness of L2 Instruction“:

Putting these principles into practice would be problematic, however, if for no other reason than that our knowledge about which aspects of language develop in a fixed order and why they do so is still too limited to make reliable pedagogical decisions (DeKeyser, 1998; R. Ellis, 1997b; Lightbown, 1998, 2000).

Therefore, the instructed SLA approach seems defeatist. But there is light. De Graff and Housen continue:

But the effectiveness of instruction need not be as limited as the claims above suggest.

There is a clear line of evidence that instruction does have an effect on language learning. Just take a look at study abroad research that compares implicit language learning (acquisition via immersion, the SLA ideal) to classroom instruction. In most studies, classroom instruction has a superior effect. Clearly, that shows instruction plays some role. How great, of course, remains to be seen. What Jordan presents as an open and shut case, that language development cannot be altered by instruction,  is actually still undergoing research.

For example, de Graff and Housen argue that the order of development may not be as immutable as it seems:

processing constraints, which may indeed be impervious to instructional intervention, the orders are primarily caused by “learner-external” features, such as
the perceptual saliency of linguistic features in the input or their communicative value (Goldschneider & DeKeyser, 2001). If so, the notion of developmental readiness is downplayed, and there may still be a stronger role for instruction if it succeeds in manipulating either the amount and saliency of exposure or the learner’s input.

There is research (as pointed out in the de Graff and Housen article) that  some instruction can help learners exert control over their developing knowledge. Some instruction can help alter the route, the speed, and the proficiency level of language learning:

the net effect of L2 instruction is substantial, so that ‘L2 instruction can be characterized as effective in its own right’ (Norris & Ortega, 2000, p. 480)

So, the order of development may not really matter. Students may be able to learn what you teach. Of course, a keyword here is some. Not all instruction is equal. There are a number of factors, as usual. Explicit instruction, including explicit focus on form instruction, is seen as a well-supported type of instruction. In addition, a number of learner variables affect the effectiveness of instruction. These include age, cognitive maturity, context of instruction (SL vs FL), aptitude, working memory capacity, proficiency, motivation, etc. We can include how salient and complex the target language is, as well. What type of instruction for which learner variable is not clear, however. Some evidence suggests, for example, that explicit instruction is better for intermediate students.  Some research has found explicit instruction better for simple structures while others have found it better for complex structures.

What this equivocal research does is complicate the entire language learning process, but it does not render it superfluous (as I have interpreted instructed SLA’s conclusion – and my interpretation may be wrong). Again SLA research, like language development, is not immutable; it is ongoing. It shows that we have insight but not irrefutability.

Back to Jordan’s arguments and the point of this post. I don’t necessarily disagree with the basic SLA principles Jordan espouses, nor many of his arguments against coursebooks and the rigid misuse of PPP. What I’m most concerned with is a static view of SLA and the fact that TBLT is being promoted as the only methodology that supports SLA even though it fails to address, just as PPP seems to, the highly personalized nature of language acquisition.

PPP assumes learners will learn what is taught, practiced, and produced. What if it follows the principles of SLA, helping to develop explicit knowledge of a highly salient feature of language that learners find meaningful and useful? What if it gives explicit practice, corrective feedback, and enough controlled and productive activities to foster implicit learning and lead to automaticity? Then, shouldn’t we expect students to learn what we are teaching them?

Neither PPP not TBLT can target developmentally readiness. But they don’t necessarily have to. While SLA research shows a firm order of development, we have learned that instruction can certainly impact it. Through understanding students wants and task needs, offering meaningful opportunities for language learning, explicit instruction, corrective feedback, working with cognitive load in mind, providing opportunities for recycling (all evidence-based principles), we can affect the route, speed, and level of language learning. With these principles in mind, both TBLT and PPP have their place.  To conclude with de Graff and Housen’s words:

Whatever the case may be, the position taken by most researchers today, including the authors of this chapter, is that SLA is a process which can be influenced by instruction, though not necessarily ad libitum, and it is exactly this relative openness of SLA to instruction which has to be explored.

Other articles about PPP


De Graaff, R., & Housen, A. (2009). Investigating the effects and effectiveness of L2 instruction. The handbook of language teaching, 726-755.

DeKeyser, R. M. (1998). Beyond focus on form: Cognitive perspectives on learning and practicing second language grammar. In C. J. Doughty & J. Williams (eds.), Focus on form in classroom second language acquisition (pp. 42–63). New York: Cambridge University Press.

Ellis, R. (1997b). Second language acquisition. Oxford: Oxford University Press.

Goldschneider, J. & DeKeyser, R. M. (2001). Explaining the “natural order of L2 morpheme acquisition” in English: A meta-analysis of multiple determinants. Language Learning 51, 1, 1–50.

Lightbown, P. (1998). The importance of timing in focus on form. In C. J. Doughty & J. Williams (eds.), Focus on form in classroom second language acquisition (pp. 177–96). Cambridge: Cambridge University Press.

Lightbown, P. (2000). Anniversary article: Classroom SLA research and second language teaching. Applied Linguistics 21, 4, 431–62.

Norris, J. & Ortega, L. (eds.) (2006). Synthesizing research on language learning and teaching.
Amsterdam/Philadelphia: John Benjamins.

Ortega (2009) “Sequences and processes in language learning”. In Long and Doughty (2009) Handbook of Language Teaching. Wiley.

VanPatten, B. (2017). Situating instructed language acquisition: facts about second language acquisition. Instructed Second Language Acquisition, 1(1), 45-60.

12 thoughts on “Is it time to SLAy PPP?

  1. Hi Anthony,

    Interesting post. As much as I do find TBLT more effective (intuitively), I do not think all PPP is bad, it just depends how the first two Ps are done and if there actually is a final P. Cheers!

    • Anthony Schmidt says:

      Thanks for the reply, Marc. At some point during a TBLT lesson, there is explicit focus on form, I believe. I can see this as a good use of PPP in TBLT.

      • Hi Anthony. Yes, there is explicit FonF in TBLT but it needn’t follow a PPP method. Long (2015; see also 2008?) prefers recasting. I prefer a written recast where possible, and not just grammar/syntax (which is usually focused on too much in both PPP and other lesson types) but pragmatics, semantics, and pronunciation/phonology. Cheers for the reply.

  2. wotcha Anthony

    thought putting here a summary of teawithbvp ep 65 would be relevant as a counter:

    instructed teaching – explicit teaching of forms (or grammar) doesn’t affect route nor rate nor extent

    instruction makes a difference in the short-term (a week, 2 weeks, at most a month) but:
    – tests tend to measure explicit knowledge
    – long term (8-9months, 1 year) gains disappear
    – sometimes instruction impedes acquisition, slows it down (3 studies only though)
    – instruction attempt to teach product (rules) not process

    what is happening is input quality matters – effects of literacy, elaborated input, speeds up acquisition or extent of acquisition. i.e. written language is more elaborated than spoken e.g. kids in L1 English don’t acquire passives until 8/9 years as passive is not common in spoken language

    one study (a Vanpatten one :/) that did show longterm effects (>9months) was processing instruction i.e. instruction on process

    what roles individual differences play are unclear;

    overall studies are messy and we cannot conclude based on this evidence that instruction makes a difference


    • Anthony Schmidt says:

      VanPatten’s article (referenced in my post) did point out similar evidence. It seems there is contradictory research. This to me is good because it indicates the conclusions are not final (i.e. we don’t know everything we need to know about language learning), and that there needs to be more students with delayed post-tests, as most know there are a dearth of these. Of course, longitudinal studies would also be great, but don’t hold your breath!

      Again, though, we also have to weigh the evidence currently available, and filter it through our biases (a natural tendency). If we think instruction has no effect on learning, then i just don’t see the point in trying. If we think instruction does have an effect, and we are inspired by the current research, then we need to keep going with the understanding that we may not be as effective as we want but we must be having some impact.

      • Note how instruction is defined in most studies – teaching of rules; this “ineffectiveness” is what VanPatten is referencing; looking at instruction as providing quality input for the language system to work on paints a more upbeat picture (if you accept the generative grammar approach of VP – which i tend to do)

        • Anthony Schmidt says:

          I belive instruction has several definitions, and rules is one which is conducive to research. Nevertheless, SLA, VanPatten, Ortega, Norris, de Graff, and others have entirely too much focus on grammar. None of the research here (SLA) looks at reading, writing, or listening nor lexical development. It seems like there is, for ease on observation, too much focus on grammar and speaking.

  3. It’s fair to argue that there’s too much focus on grammar & speaking if you want to argue for more attention to other things, but I’m sure you’ll agree that the effects of instruction on vocab. learning, collocations, lexical chunks, reading, writing have been studied.

  4. Interesting post. Did you look at the other article in that first issue of Instructed SLA?

    In it, Long argues “the ultimate purpose of ISLA research is to improve L2 […] learning or teaching, and to do that, some connection with language acquisition is required.” It’s true that he spends most of the paper talking to researchers, setting out a research agenda, and citing some studies which he considers valid and reliable. I understand his argument as being that the field is too young and/or SLA too complex for concrete recommendations to teachers, but his view of the purpose of L2 research seems to match your concerns:

    “A major goal of ISLA research is to free instruction of unnecessary artificial aids, unless they turn out to be either absolutely necessary or an improvement, e.g. because they produce implicit knowledge faster.”

    I’ve just been watching Widdowson’s BAAL 2017 plenary where he is somewhat critical of this approach, though he also attacks the CEFR and its functional approach to communicative competence. But I think he also speaks to some of the concerns you raise here

Comments are closed.