Last changed 7 Feb 2003 ............... Length about 4,000 words (23,000 bytes).
This is a WWW document by Steve Draper, installed at You may copy it. How to refer to it.

Vocabulary as a barrier to learning

Stephen W. Draper

Contents (click to jump to a section)

Short overview note

This document is a note about the issue of vocabulary in learning (in HE), based on a message I sent to the ITFORUM email list in December 1997. (Link to list of ITFORUM participants.)

Vocabulary (jargon) has been shown in research by Alex Johnstone here at Glasgow to be a significant barrier to learning, at least in this university. Furthermore, teachers here don't care ("my job is to teach Chemistry, not English"). Alex has convinced me of this, but I am having trouble developing a systematic solution in my practice (i.e. in taking David Tai's advice to practice somebody else's sermon). In the current discussion someone said that learning the jargon of a field is an essential part of getting educated. That is true, but few teachers take that point, not just as a response to student complaints, but as a learning objective they need to address by separate actions: e.g. beginning EVERY course by handing out a glossary as a MINIMUM towards fulfilling this requirement; and for instructional designers, ditto.

Part of my problem is that it is not entirely easy to discover what needs to be in the glossary. I ask myself what is clear, but my intuitions are just too different from my students' to be at all reliable. I ask my students, but they seldom tell me, at least in response to a bald question. The most deadly cases, as Alex showed, are not in fact overt jargon like "mathemagenic" which at least every student immediately knows they don't understand, but ordinary words used technically ("work" in physics, "volatile" in chemistry, "reliable" in psychology, "active learning" here). Everyone thinks they understand, but often they don't understand the same things.

It is impossible in practice for communicators to predict as they write how they will be misunderstood, although you can certainly make a damning-looking case after the event. Larry was just using the vocabulary he was most comfortable with, and doubtless uses all the time in the company he feels most at home in. So do teachers and contributors to this forum, but it doesn't lead to reliable communication.

Practical remedies

Asking learners doesn't work very easily, not least because the worst problems are the words the learner thinks they understand, but don't because they have both a technical and an everyday meaning.

This is one of the (few) things the WWW could be a real solution for: we can collect terms, we can continually update it, as this problem doesn't stand still, and we can share a glossary across many institutions as it isn't very sensitive to having extra terms one particular user doesn't want.


Cassels, J.R.T. (1980) Language and thinking in science — some investigations with multiple choice questions Ph.D. thesis, University of Glasgow

Cassels, J.R.T. & Johnstone, A.H. (1983) "The meaning of words and the teaching of chemistry" Education in chemistry vol.20 pp.10-11

Cassels, J.R.T. & Johnstone, A.H. (1984) "The effect of language on student performance on multiple choice tests in chemistry" Journal of Chemical Education vol.61 pp.613-615

The Royal Society of Chemistry (1985) Words that matter in science

Johnstone, A.H. (1991) "Why is science difficult to learn? Things are seldom what they seem" Journal of computer assisted learning vol.7 no.2 pp.75-83

Skemp,R.R. (1976) "Relational understanding and instrumental understanding" Mathematics Teaching issue 77 pp.20-26

An experience and terms for fields

This is a note I wrote on 1 Feb 1998

I recently had an experience I am still digesting concerning vocabulary in teaching.

I ended a block of 6 lectures to a final year undergraduate class by putting up an OHP slide with some of the long words that had been used, and asking the class what they meant. I did this because of the research showing that a failure of common understanding between teachers and students of technical vocabulary is a major barrier to effective teaching in my, and probably in all, universities. It felt a comfortable thing to do, both in planning and in execution. The students seemed to say it was useful to them, although one said I should do it at the start as well as at the end of a block of lectures. For technical words I had introduced, this was a good concluding tactic. For some other words it seemed to show some of the worrying problems the literature had warned me about. I also realised that I was as hesitant and inconclusive as the students about actually defining some of the terms.

By serendipity (or anyway coincidence), only an hour or two after that I read some accumulated emails on ITforum about how to define the field of Instructional Technology, in which one person argued that there is not and cannot be a sensible definition for names of academic fields: they have a developmental history and such words are necessarily like Wittgenstein's example of the word "game". They don't have definitions specifying necessary and sufficient conditions.

I am still wondering what to make of this. From research literature I know that vocabulary is a serious problem, and that it is ignored by almost all university teachers, but that this damages teaching and I should be doing something about it in my own practice. I certainly feel that I should be able to define the words I use to my students. But the argument says I shouldn't. And in fact it is true that the terms that had us all stumbling the most referred to fields: neurophysiology, psychophysics. And it must matter. For instance I recently used the word "neurophysiology" in an exam question. What reason do I have for believing the students knew what it meant? I can't remember telling them. And if they didn't know, then the whole question reduces to a trivial surface test of vocabulary.

What should I make of this?

And mind you, all the above is about overtly technical vocabulary. But the most deadly problems are not words like "phenomenography" and "mathemagenic" which everyone instantly recognises as a problem and may ask a question about, but words with a familiar everyday meaning different from the technical one(s) such as "derivative", "volatile", "subject", "cell", "attitude"; and probably also "learning", "memory", "vision", "perception": not just "valid", "significant", and "reliable". The deep problem is how to know which of the words I used conceals a misunderstanding neither I nor my students recognise.

Skemp (1976) introduces his paper (which is mainly about kinds of understanding in maths) by discussing the French term "faux amis" for words that look the same in two languages, but actually have seriously different meanings e.g. "chef" in English (means cook) in French (means the head of any organization). This, in a cross-language context, is the issue.

Vocab: working summary


To collect provisional thoughts on the vocab. in teaching problem; particularly what to do about it.

Note too that the situativity view of learning and teaching could be regarded as saying that vocabulary is the whole of education: learning to talk the talk in order to be part of the community.

The problem

The overall problem is words used by teachers that learners do not in reality understand.

Learners detecting a problem

The associated problem is whether the learner can detect that they do not understand it. If they do not, they will not even try to discover its meaning. In relation to the latter, there are:
  • A. Jargon: unambiguously not-understood words e.g. "mathemagenic". Students immediately know they do not understand, and may therefore ask.
  • B. Ambiguous words, deceiving learners by having a meaning they know, but that is different from the technical meaning actually intended by their teacher. Learners cannot know they do not understand, and so cannot ask.

    Relationship to teachers' goals

    Problems fall into these broad classes, differing in how likely a teacher is to predict a likely problem and provide an explanation without being asked:
    1. Technical words central to the current teaching. Mostly the teacher will explain their meaning: they are part of the content to be taught.
      • Obvious jargon e.g. mathemagenic
      • Hidden jargon e.g. valid (a technical word in psychology).

    2. Technical words not the focus but used by teacher
      • Names of other fields e.g. neurophysiology, ...
      • Asymptotic, Euclidean geometry, acronyms, iteration.

    3. Educational / institutional terms. These words are often used by teachers in directives to students, yet a) usually no effort is made by anyone to define their meaning for learners, b) frequently different teachers in the same institution use and assume different meanings.
      • Organisational terms: Seminar, tutorial, semester;
      • Terms about work e.g. in an essay specification: Argue, discuss, bibliography

    4. Supposedly common English words, not in fact correctly understood by the learners.
      • [Martie Sanders' examples] simultaneous, abundant, subsequent
      • Logical connectives: therefore, however, because, thus (and probably: if, then).

    Why should teachers address the problem?

  • Communication. Part of teaching is communicating with students. This involves negotiating a common language: actually a sublanguage both understand. Taking the view that they should know the teacher's language before they come is to refuse to undertake a standard part of ordinary human communication, let alone teaching. (On the other hand, it isn't all that easy in a large class to discover and adjust to misunderstandings, which in one to one communication you would see from a raised eyebrow, and so on.)

  • Technical vocabulary is a learning objective. All subjects have their own vocabulary, and learning the subject is in part learning that technical vocabulary. So on the one hand, teachers shouldn't just talk down in a non-technical vocabulary, but on the other hand, teaching that vocabulary should be seen as an explicit learning objective in every course; written up as such, and normally given separate teaching actions.

  • General vocabulary is a learning objective. Vocabulary that should be part of an educated person's repertoire; that is, a general educational objective.

    Definitions? philosophy

    A view deriving from the later Wittgenstein is that words do not necessarily have precise definitions (necessary and sufficient conditions) e.g. "game". Words only need to have usages in context to have meaning. We understand each other by other means than consulting knowledge of meanings like definitions, resting ultimately on our shared experience of and position in the world.

    However in the context of teaching, at least of teaching any technical subject, I think I have to confront such general vague philosophy with some realities. Here, in a microcosm to do with classroom needs, is the Wittgenstein-ian debate between formal ideas of language modelled on axiomatic logics and ideas of language as an interaction between beings whose use of words depends upon a shared meaning based on something else.

    Whatever the partial truth of that, at least the following opposing facts must be accommodated in our approach to learning and teaching:

    I can only say that teachers should provide explicit definitions and examples of terms wherever possible; and in other cases indirect definitions and examples; and failing even that, provide examples and pointers to where learners can read further examples.

    Things to try

    1. Start course sections with vocab?
    2. Keep word list on blackboard as you go, during a lecture. And on the web during a course.
    3. Close course sections by discussion of vocabulary e.g. bring an OHP with it on, and discuss or ask students about each word in turn. This is probably a good idea, but likely to focus only on the words the teacher is conscious of being a problem.
    4. Web glossary built by students.
    5. Produce a glossary for every course or lecture section (perhaps on the WWW). As it is a learning objective, even if often tacit, it should be done. Textbooks with glossaries are usually not satisfactory as work at Stafford has shown: if a student needs to learn (or check) say 30 words as part of a block of lectures, it is not useful to have a glossary for a whole degree course or profession with a thousand terms in, hiding the few they need. We should probably produce a vocabulary list for each small unit of teaching. Putting it on the web may be a particularly good idea, as it can easily be updated as more problematic terms are exposed.
    6. Draft a document and go round getting definitions from staff, or at least asking for comments on your own definition. This is particularly important for type 3 above: institutional terms that in fact may not be being used consistently.

    How to get students to ask about words; and to deal with that effectively during class.

    Context and audience: towards a design method

    David Nelson is interested in developing a design theory for correcting vocabulary useage in instructional materials by concentrating (the designer's mind) on a) context b) audience. Here is a message I wrote to him about this.

    You want to create for vocabulary a model as practically useful as ARCS for motivation (or as Laurillard's for pedagogic/instructional design, as being developed by Michelle Montgomery).

    To do that, you have to find a way so that, given proper study and practice of your model-to-be, any teacher could review and modify their text (or other material). The difficulty I feel is that this will require knowledge of what students whom you haven't yet talked to know. Perhaps I shouldn't be so resistant. Perhaps we (teachers) should get serious; look at the required educational pre-requisites for students taking this course, and really just read off what they know; anything else they can't be assumed to know.

    So for my "asymptote" example, after titanic recall, I think that term comes with learning about hyperbolae in the math of conic sections. So I would "just" need to look at current math syllabi (syllabuses?) to see if that is still taught at school as it was for me; and whether my students would have taken math A-level (advanced level high school math, roughly).

    I teach HCI (human computer interaction), and this goes much better (like most things) using real examples from the audience's personal experience. On the course I teach it in, this means I need to familiarise myself with their current computational environment. This is some extra work, but guarantees the relevance of my examples for the whole class. This is a close parallel to, though perhaps rather easier than, what we are discussing here.

    If you continue to be serious about this, here are 2 extra related considerations. You said that page turning isn't real hypertext in the context of using hypertext for vocab. explaining. While I agree there is probably a factor of about 10 in speed of access, so in one way your point is right, on the other hand you should consider carefully the real use of such vocab. lookup. I didn't mention any of this in my messages, but of course the reality is that in many cases we feel we understand enough from context not to need instantly to know an exact meaning for words. I don't understand when this is and isn't true. But when it is, slow page-turning lookup may be OK.

    For me, this is related to Beatrix Potter. She is still an ultra-famous English children's author, writing about 100 years ago for the very young. I read in a biography that she was criticised by some at the time for including some rare and long words completely unsuited to such a young audience. She rejected this criticism, saying something to the effect that to use an infantile vocabularly rigidly was patronising and prevented their learning new words. I've recently heard comparable remarks in feedback from students: something about positive attitudes to a professor here who presupposed wide reading on their part. This was completely unrealistic, but made them feel they were being taken seriously and indicating a standard to which they could aspire. There is a very important point here, though I haven't got on top of it yet.

    Perhaps this is what you mean by "context". But it is to do with how some words you just don't need to understand immediately; others you only need an approximate idea at the time; but others are essential or you are lost from then on.

    I suppose it may turn out something like:

    But there will be additional principles.
    A difference between our PSA and asymptote examples is probably that your students would be enriched by hearing a 3 minute diversion explaining all about what they are (say because designing PSAs and thinking about their effectiveness or lack of it is an interesting ISD sideline); but mine would probably not be as conic sections are totally unrelated to HCI and all I needed was a way of talking about a limit that you forever approach but never reach. So your model-to-be should not discourage such things, but instead require teachers to estimate whether it is a case of a useless extra burden or a Beatrix Potter-like case of expanding their experience beyond their immediate pedagogic need but not beyond any relationship to that.