This post will be quite long, and probably boring to many, so fair warning.
Anybody who is an academic, particularly among those who are young and looking to make a name for themselves, is likely to feel frustrated by the pace of knowledge generation and the constant need to qualify and contextualized claims. I wouldn’t have it any other way; this is how we’ve responsibly built knowledge for centuries and it works. It is frustrating, sometimes, to constantly put a “we think” when you want to put a “we know.” It’s a temptation I have to fight myself a lot.
Well, one thing that I feel comfortable saying we just know in writing pedagogy is that teaching grammar doesn’t work. And I think grammar is a useful lens through which to examine larger issues in education.
A vast swath of empirical research demonstrates that teaching grammar is useless. A great review of the literature and appropriate citations can be found in Constance Weaver’s Teaching Grammar in Context. (As you’d imagine, that book does concern a certain kind of directed grammar pedagogy, but in a very different manner than grammar instruction is typically understood. More on that in a bit.) You can run down a laundry list of common claims about the virtues of teaching grammar that have been undermined by empirical study.
- Grammar instruction does not lead to better reading outcomes
- Grammar instruction does not lead to better global writing outcomes, in terms of producing better reasoned, better supported, more coherent, better organized, and more stylistically mature writing
- Grammar instruction does not lead to better local writing outcomes, in terms of producing better mechanics and fewer surface errors
- Contrary to long-standing assumptions, grammar instruction has not been shown to produce improvements in mental discipline, study habits, memorization or other meta-educational areas
- Despite early optimism, the switch from teaching traditional grammar to teaching structural linguistics has not shown improved student outcomes
- Neither has the switch to transformational grammar
Funniest/saddest of all, there’s evidence that teaching grammar in the traditional sense of sentence diagramming and parts of speech fails to improve student outcomes on tests of… sentence diagramming and parts of speech. That is to say, grammar instruction seems to fail even in the most artificial context of learning grammatical terms for their own sake.
There are, of course, dissenting voices and qualifications, but the failures of grammar instruction have been demonstrated over decades of research undertaken by dozens of researchers in a wide variety of contexts examining students from a large array of ages and backgrounds.
There’s a lot of rationales that you hear for why grammar instruction seems like such a pointless endeavor. I find most compelling the notion that learning grammar is difficult because learning functional grammar is the work of language acquisition, at once the most important and least directed or controllable aspect of human learning. No (well, very few) parents show their infants flash cards and practice sentence construction in front of them. Grammar learns you. We should remember that what can be taught and what can be learned are asymmetrical categories.
It’s important to note, though, that though functional grammar acquisition happens without teaching, that doesn’t mean there aren’t students with better or worse grasps on effective language use. And after all, some people do know parts of speech, diagramming, and other aspects of grammatical nomenclature and theory. I’m sure many of the people reading this do. Speaking purely from anecdote, my own experience is that the people who learn descriptive and prescriptive grammar are those who already have facility in functional grammar. They know how to effectively construct English texts first and then learn the nomenclature and diagramming later. Which is fine for them, but nearly useless for us as educators, as our goal is not to use functional grammar to teach descriptive and prescriptive grammar but the other way around. (It would take a truly old school sensibility to think that learning what a gerund is has value in and of itself.) It’s a common theme in education: the rich get richer, while those who are behind fall further behind.
If I could put a gloss on the whole situation, I’d say that the misery appears to be that a certain degree of syntactic maturity is necessary for effective teaching of the writing skills that are so important for success in college and the workplace, but it is very difficult to impart that syntactic maturity to those who did not acquire it from their environment early in life.
You can imagine what the difficulty of this situation is for teachers of writing like me. We have very compelling data that suggests that we shouldn’t bother trying to teach grammar. But we also sometimes have students that lack the elementary sentence and paragraph construction skills (which for ease of use we’ll refer to as grammar) necessary to move to the higher-order, global composition concerns that are the work of a collegiate writing classroom. The sources of this deficiency are varied and complex. There’s a long conversation you could have about remediation, high school writing pedagogy, TOEFL and ESL, and so on. But once these students are in our classrooms, the issue is moot; it is our responsibility to teach them. So the question becomes how to try and teach them elementary language structures when they lack the grammatical nomenclature necessary to be taught and when a mountain of evidence says there is no utility in trying to teach that nomenclature. This is an issue that I discuss constantly with my professors and my peers.
Currently, I and many others are muddling through. I do at times teach a kind of small-bore, usage specific grammar in my classes. I identify mechanical problems that are repeatedly appearing in the work my students give me, demonstrate the problem in class, and show them how to avoid the problem in the future, up to and including giving them sample constructions and models for correct usage. There’s no nomenclature beyond the bare minimum necessary to have the conversation. There’s no diagramming. There’s no attempt to tie the immediate issue to broad grammatical categories.
I have seen some middling improvement in subsequent papers. In conversation with other instructors I’ve found that this is not unique to me. But I have to be honest: I’m unaware of responsibly generated empirical data to support what I’m doing, and while the tight focus of our grammar workshops has its virtues, it also means that I’m not systematically addressing broad syntactic deficiencies. I’m not confident that these lessons are adding that much value, but I also can’t proceed without addressing these concerns. For now, I keep muddling.
All that wind up, what’s the pitch for the broader themes in education?
- Students may all be made equally, but by the time they reach educators, they are not equal. I’m consistently amazed by the prevalence of thoroughly discredited blank-slatism expressed in the popular press. Educators are constrained by what students are bringing with them into their classrooms. I have a lot of faith in the ability of our educational psychiatrists, neuroscience, and linguists to eventually figure out exactly how and when functional grammar is absorbed. I have less faith that there will be an obvious policy fix that emerges from that research, given socioeconomic realities and parental rights.
- There are limits to what can be formally taught. My old saw: the fact that people in the media and policy apparatus really want it to be true that all students can be taught everything is not an argument that they all students can be taught everything. In the context of our education debates, statements like “grammar cannot be effectively taught to a majority of students” tend to get people riled up and talking about our “can’t do” teachers unions, etc. etc. etc. But reality is just indifferent to both anti-educator axe grinding and cheerful liberal daydream believing. We’ll keep working on functional grammar instruction, but commitment doesn’t ensure results. It never has.
- Disadvantage in student environment and disadvantage in schools are separate issues. This might seem a bit far afield of what I’ve been talking about, but I think it’s related and crucial. If (if) the evidence says that language acquisition in the period before formal schooling partially determines someone’s later facility with language, then it makes sense to consider the environment in which that person is growing up. In that context, I think it is very important not to make a common elision between conditions in the home and conditions in schools. Often people will talk about higher school funding as a solution to student poverty effects. Better funding for schools is important and I support it, but the average K-12 student spends less than a fifth of his or her life at school. The lack of an environment conducive to the acquisition of functional language skills is precisely the kind of problem that can’t be solved at the school level and points to larger structural social problems.
I’m not being a nihilist here. There are plenty of things we can teach. I have been accused of being overly pessimistic about educational outcomes, that I am saying that we just can’t teach some people and should stop trying. This would be a weird conviction, given that I am quite literally dedicating my life to improving our teaching of writing, and in particular for those who have traditionally been worst served. I have a considerable personal, professional, and academic commitment to improving educational outcomes in a particular slice of the pedagogical project.
Pedagogical innovation is real, and sadly under-discussed in our educational discourse. I believe strongly that I’m improving the skills of even the most marginal students that I teach. And once students have acquired the functional grammar necessary to move to higher order concerns, there’s a lot of effective pedagogical tools we can use. But saying that we are helping all students and saying that they are all reaching the level of proficiency necessary to succeed in college or the working world are very different things. People invested in these issues simply have to take the idea that there are limits to what educators can do for individual students more seriously, and we have to recognize that while education is an important lever for creating a more just and equitable society, it is not and can not be the only one. Trying to make it so will only cause more people to fall behind.
In an environment where we are constantly confronted by the fact that we can’t bring all students to the same minimal levels of competency, despite how badly we want to, we have to look to more direct measures to ensure an egalitarian society. You solve for the fact that not all people are equally equipped for success in an information economy, due almost entirely to factors beyond their control, by building a far more economically redistributive society.