Project-based Learning in the Second Language Classroom

Project-based learning is a staple in my Social Justice, English and Philosophy classes.  When I heard that it has been tried in the language classroom, I was curious.  I read a detailed lesson plan for a Spanish unit and thought about it.

So here are some observations on this unit, whose objective is to get students– who have decided on (or come up with) a new invention “to convince a Spanish-speaking audience that they need this new invention.”  These are low-mid and late novice students.

We know from research and from successful (i.e. significant and positive results-based) teaching for language acquisition that best practices include:

• a focus on high-frequency vocabulary (yes, this varies somewhat by context)
• keeping input comprehensible and vocab limited
• little focus on output
• a lot of input (listening and reading)
• avoidance of L1 as much as possible (outside of keeping L2 meaning clear)
• providing quality input (ie, not what beginners inevitably produce)
• reading and listening to “whole” language (stories, dialogues, reports etc  etc, not lists or discontinuous text) etc.

So how does P.B.L. stack up against input-based language-teaching  practice?

What I first noticed is the heavy emphasis on early output and what we would loosely call “communicative pair activities” (CPAs).  There is a lot of “learn to say ____” and “get someone to respond by _____,” in everything from choosing inventions to forming groups.  I also noticed a lot of group work (decide, choose, evaluate, predict the meaning of ___, etc).  The aim— using the target language– is laudable. The problems here are that

a. you can expect lots of English use during these activities, inevitably.  Why would someone use L2 if L1 is easier.  We’ll expect a lot of English also because for most of these activities the students themselves have to find the vocabulary they want to use.  Since they don’t have that vocabulary, they will use L1.

b. the Spanish output— because it comes from beginners– will be limited, error-filled and impoverished.

Second, the kids go through a list of cognates, predicting meaning, then doing various activities to check the meaning.  This seems useful,  since kids certainly do need to be taught reading skills (e.g. to look for and make educated guesses about cognates).  But this takes a lot of time, and will be accompanied with a lot of English (“what do you think vender means?  –I think…”). The issues: group work means English use, and CPAs mean junky Spanish.  If language is acquired through input, I don’t see how poor input and English are helping the kids.

The implicit claim here is, predicting and then checking meaning is a helpful learning strategy.  This is true– for explicit (conscious) learning.  For language, not so much. What we need to acquire language is comprehended input.  That’s what we give our learners, and what parents give their kids. Yes, kids will do some guessing about vocab when acquiring their L1s, but this is OK for them, because they have 100s of hours.  We have a couple of hundred.

Here is one of the activities:

Third, an issue here is the use of word lists.  Language is not acquired through memorising, practising or predicting/checking the meaning of words on lists.  The brain is pre-wired to process whole language:  meaningful sentences which are part of bigger communicative acts (stories, conversations, etc). You can memorise via lists…but it’s not fun (read: many kids won’t do it), and it’s not efficient.

Another problem with lists: the “grammar” that “ties together” words is absent. Lists provide impoverished input.

Fourth, we have the problem of the use of low-frequency vocabulary.  Here are some examples from the unit. These are necessary words for marketing.  However, according to the Wiktionary Spanish frequency list, none of these words are in the 2000 most-used Spanish words.  Why teach this to novice or mid students?  If you want to get your kids ready for Mexico or Spain (or the A.P. exam), you– they– are much better off  reading a ton of writing which full of high frequency vocabulary, i.e. the stuff they are going to actually hear a lot in Spain and Mexico.

Fifth, we have the interest and learning problem.  If the class is going to share these projects– i.e. the kids present their projects to other kids– how are the various groups going to learn each others’ vocab?  Viewers of reports, presentations, posters etc will get short-term and non-repated exposure to the vocabulary.  And why should they necessarily care about others’ work?  If the point of P.B.L. is for students to engage in what interests them, why should they sit and listen to what others have done? If I like sewing, and Suzie likes motorcycles, why on Earth would we want to listen to each other?   This problem is solved– to the extent that it can be– in a TPRS, narrative paraphrase, A.I.M., Story Listening or other storytelling classroom, where stories and characters (which are almost universally interesting) are the focus.

Sixth, we don’t have any evidence (of which I am aware) that PBL works from a proficiency-oriented basis. That is, PBL advocates have not shown us what kids can do (without notes, dictionaries etc) after lessons.

Blaine Ray has said that the litmus test of successful language teaching is the unannounced, timed write, where students get, say, ten minutes to describe a picture, event, person etc without any advance notice, preparation, use of notes, etc. This evaluates acquisition— what people have “wired into” their heads. Lots of C.I. teachers share results. I do. Adriana Ramírez does. Online groups regularly do. Meredith White does.

PBL advocates– as far as I know– havn’t shown us proficiency-based, zero-prep results. The PBL kids however do do quite well when they have lots of access to notes, when they have time to prep, memorise presentations, etc.

So…from what I have seen, P.B.L. is not going to fit in with comprehensible input-based instruction.  Caveats:

• I havn’t used it, and so I don’t have any data to support my gut feeling that P.B.L.’s effectiveness is limited.
• Maybe you could do PBL with shared vocabulary, ie the whole class decides on a problem, and gets some strict guidelines about what can/cannot be used.
• Bill VanPatten has commented that “PBL is not an appropriate teaching strategy for most language learners,” because “they don’t yet have enough language in their heads.”
• You could easily make PBL work in L1.