Skip to content

Teach the conceptual underpinnings of L2 forms rather than only their structural aspects or formulaic units

Principle · All principles
Evidence · 19 claims (10 for, 7 mixed, 2 against) · 16 studies (5 theoretical, 3 causal, 3 quant-synthesis, 2 review, 2 associational, 1 qualitative), q1–q3 · 0 of 16 report an effect size · 16 claims rest on one study

Conditional relationship

This page covers explaining the meaning behind a second-language form so that learners choose it in the right contexts. The relationship it proposes is conditional. For a second-language learner (usually intermediate or advanced in the L2, usually adult) who already produces a form, such as a preposition, a discourse connective or an aspect or article choice, but uses it in a pattern unlike that of proficient users (overusing it, avoiding it, or extending it to contexts where it does not fit) because the L2 divides up the meaning differently from the learner's first language, an explicit explanation of the meaning that motivates the form's uses (its core sense, and how each further use extends from that sense, shown in a diagram or schema), followed by practice in choosing the form across contexts, is expected to produce more appropriate choices (concept knowledge of where the form fits, and skill-performance in the learner's own comprehension and production) than presenting only the form's structure or teaching its uses as unconnected fixed phrases. The expected change is at an immediate horizon on researcher-designed measures; use in the learner's own unplanned speech and writing, and delayed-retention, are separate claims.

It is conditional on four things: that the form's uses do have a describable meaning motivation (for prepositions this is the cognitive-linguistic account of a spatial core sense; for an arbitrary agreement rule there is no such story to tell); that the learner's difficulty is one of distribution, not of not knowing the form; that the first language's nearest equivalent covers a different range of meaning; and that the learner can follow an explanation about meaning, in the L2 or the L1.

No claim in the wiki tests this relationship. The one reported study (Advanced L2 learners taught the core senses of to, for and at with diagrams show pre- to post-test gains in knowing when to use them, in one uncontrolled study reported second-hand [+M]) is described second-hand in a theoretical article, as a pre-test to post-test gain for advanced learners taught the core senses of to, for and at with diagrams; the claim page reports no comparison group and no effect size, so it does not compare conceptual explanation with structural or formulaic teaching. The nearest comparative evidence is the larger literature on explicit against implicit grammar instruction (Explicit L2 grammar instruction leads to larger learning effects than implicit instruction, as reported in the reviewed synthesis literature [+S]), which supports making a form explicit but does not say which kind of explicit explanation, conceptual or structural, does better.

How the neighbouring converted pages sit beside this one. Error Analysis holds learners studying errors set beside correct examples, and records that the teacher's analysis of learner language is a diagnostic method with stated limits; this page uses that kind of analysis only to choose which form to teach, and does not repeat its model. Explicit Instruction: Vocabulary holds teaching the meanings of selected words; this page is about grammatical and function forms whose many uses are linked by one meaning. The Concept Attainment pattern gives a sequence of examples and non-examples that can carry the choice practice below. The recipe that applies this page to prepositions is the strategy core sense first, with diagrams; the element Concept-Based Language Instruction with Gal'perin's verbalization phases supplies the step where learners explain the concept aloud.

Default design, while the relationship is untested

The earlier page's guidance (identify the conceptual underpinnings and core senses, teach them explicitly, aim at appropriate use of forms whose distribution differs subtly from the L1), made concrete. Each step is a design proposal with its evidence status beside it.

  1. Confirm that the problem is distribution, and for which form. Collect a sample of 15–20 of each learner's uses of the candidate form from their own writing or speech, plus a short comprehension task, and tally overuse, misuse and absence against a reference (a proficient-writer corpus of the same genre, or a teacher's judgement, stated as such). Choose forms where learners' choices deviate from proficient use, and spend class time there. Basis: Awareness of how learners' conceptual representations deviate from native-speaker preferences can help teachers decide where to focus classroom attention, promoting conceptual fluency [+W] (an attributed proposal, no data). Include the comprehension task and an elicitation that requires the form, because a form a learner avoids shows no errors (Error analysis has stated limitations: overattention to errors, overstressing production data, failure to account for avoidance, and language-specific focus [~W]). The sample size is an untested proposal.
  2. Set the L1 and L2 meanings side by side. For the chosen form, write down which contexts the L1's nearest equivalent covers and which the L2 form covers, and mark where they part. Basis: Each language entails a conceptually different world, so exact translation of conceptual structure between languages is impossible [~W] (conceptual analysis of prepositions across four languages; theoretical) and Thai students overuse the causal connective because, a pattern the article attributes to L1 influence [+W] (a corpus pattern, with the L1 cause the authors' attribution). Do not assume an L1 cause before checking it: in one judgement study, learners placed English manner adverbs correctly although Chinese places them differently (no first-language transfer for manner adverbs [~W]). The contrast step is an untested proposal; the strategy Address Mother Tongue Interference is a recipe for it.
  3. Present the core sense first, with a diagram. Show the form's central meaning as a picture of a scene (for at, a point of location; for to, movement towards a goal), then derive each further use from it, two or three examples per use, with the link to the core sense said aloud. About 10–15 minutes for the core sense and the first two or three uses. Basis: Advanced L2 learners taught the core senses of to, for and at with diagrams show pre- to post-test gains in knowing when to use them, in one uncontrolled study reported second-hand [+M] (advanced learners, prepositions, pre-to-post, no comparison group). Timings are untested proposals.
  4. Have learners explain the concept. Each learner explains to a partner why the form fits each example, using the diagram, then explains it to themselves in writing for one new example (about 5–10 minutes). Untested here; this is the verbalization phase of Concept-Based Language Instruction, which carries no claim in the wiki.
  5. Practise choosing, not only producing. Give items where the form competes with its near neighbours (to/for/at; because/since/so/as), including contexts where the form does not fit, and ask for the choice and a one-line reason. Practise in the skill the target uses: if learners must write it, they practise writing it; if they must understand it, they practise understanding it, because grammar practice has been reported to improve only the skill practised (Practice of grammar rules improves only the practiced skill, showing skill specificity of L2 automatization [~M]). Expect a low-salience form to need much more practice than one explanation suggests (Low-frequency or low-salience L2 elements are hard or impossible to learn through mere exposure and require extensive practice to proceduralize form-meaning mappings [~W]). About 15–20 minutes per session; untested.
  6. Use the form in a meaning-focused task, and give feedback that names the meaning. Learners write or speak in a task that needs the form (an argument, a set of directions, a report). Feedback on a misused form points back to the core sense ("this is a goal, not a location: which form shows a goal?") rather than only marking it wrong. Untested proposal; the principle Integrate Explicit Correction with Meaningful Activities is the neighbouring guidance.
  7. Re-sample later. Two to four weeks on, take a fresh sample of uses from learners' own work and tally it as in step 1. An explicit-teaching gain on one form has been reported not to last over months when the syllabus taught structures separately and never brought them back (Explicit teaching is more effective for EFL copular be learning but its effect is not durable [~M]). The interval is an untested proposal.

Dose and progression across sessions (an untested default for one form with several uses, over three to five sessions of about 45 minutes):

  • Session 1: the diagnostic task (10 minutes), the core sense and two or three uses (15), learners' explanations (10), a short choice set (10). The teacher draws the diagram and supplies the examples.
  • Sessions 2–3: a 5-minute recap in which a learner, not the teacher, explains the diagram; one or two further uses per session (10); choice practice (20); one short meaning-focused task (10). Examples now come partly from texts the learners are reading, which they sort by use.
  • Sessions 4–5: the diagram is no longer on display; examples come from learners' own drafts; practice moves into whole tasks; a learner explains a choice only when asked or when it is wrong.
  • Moving on: a learner (or class) moves to the next form when a fresh sample from their own work is close to the reference distribution, within a tolerance the teacher sets and records, and the learner can say why the form fits in about four of five new contexts. Learners still far off stay with that form, with examples from their own writing. Both thresholds are untested.

When a condition is not met: if the form has no describable meaning motivation, teach the pattern as a rule or as fixed phrases and do not invent a conceptual story; if the learner does not yet produce the form at all, teach high-frequency uses first and come back to the core sense once there are several uses to connect; if the sample shows no deviation, do not teach the form.

Fitting the design to a situation

The three facts that most change the decision:

  • What the learner's use of the form actually looks like. Overuse, avoidance and misuse each call for different work, and a learner who does not yet produce the form needs something else. If unstated, ask: in a sample of the learners' own writing or speech, how often do they use the form, where, and where do they avoid it?
  • The learners' first language, and whether its nearest equivalent covers a different range. That decides whether a side-by-side contrast helps, and whether one contrast serves the whole group. If unstated, ask: what L1s are in the group, and what does each use where the L2 uses this form?
  • The use that counts. Choosing the form in one's own unplanned writing or speech, passing a gap-fill, and understanding it in reading call for different practice. If unstated, ask: where, and under what time pressure, must the learner use the form, and how will that be checked?
If the brief says… Then change… Basis
Learners are novice in the L2, or do not yet produce the form Teach two or three high-frequency uses as phrases first, in meaningful use; bring in the core-sense diagram only once learners have several uses to connect, and keep it short untested; the one study (core senses with diagrams [~M]) taught advanced learners only
Learners are advanced, with long-standing errors on a form This is the main case: spend the time on the diagnostic sample and the L1 contrast, take examples from learners' own work, and ask learners to notice their own uses rather than relying on paraphrase to avoid the form core senses with diagrams [+M] (advanced learners); L1-induced errors persist in advanced learners and risk fossilization if learners rely on compensatory communication strategies [+W] (one learner's diary)
Learners are child Fewer verbal explanations: act out the core scene (stand at the door, walk to it), use pictures, and keep explanations in the language the children understand best; short choice games rather than written reasons untested; no claim here concerns children
Learners share no single L1 Do the contrast step in L1 groups with a prepared table per L1, or contrast the form with its L2 neighbours instead; expect different groups to deviate in different directions Thai students overuse the causal connective because [+W] (Thai learners overused because while Hong Kong and Singaporean learners underused it); the grouping is untested
The form's uses have no meaning motivation (an arbitrary agreement, a fixed idiom) Teach it as a rule with practice, or as a fixed phrase; do not invent a core sense untested; from the earlier page's scope ("structures with complex distributional patterns")
The target is comprehension (reading, listening), not production Make the choice practice comprehension tasks: interpret what the form signals in a sentence, choose between two readings Practice of grammar rules improves only the practiced skill [~M] (a review reporting one longitudinal study; carried from rule practice to concept teaching)
The target is fluent speech under time pressure After the explanation, move quickly to timed and oral choice practice; do not ask learners to check each form consciously while speaking, and judge progress on unedited speech Monitoring lowers accuracy for unlearnable rules [~W] (one Korean student was less accurate with prepositions in edited writing); untested as a design rule
online-self-paced, or phone-only Give the diagram as an image card the learner can reopen; make choice items automatically checked, with each explanation pointing to the core sense; replace partner explanation with a typed or recorded one-sentence reason untested
Some learners remote, some in the room Put the diagram on the shared screen and in a file everyone holds; pair remote learners with each other for the explanation step (breakout or chat); collect choice answers through one poll so both groups are seen untested
One teacher with a large class Take the diagnostic sample as one shared short writing task, tally the commonest deviation for the class, and teach that first; use mini-whiteboards or a poll for the choice items conceptual deviations guide where to focus [+W]; the class-scale tally is untested
Scarce facilitator time (a volunteer, a tutor checking in briefly) Learners tally their own uses with a checklist built from the diagram; the tutor checks three of each learner's explanations per visit and the one form the tally flags untested
A single session only (about 45 minutes) 5 minutes: diagnose from five sentences; 15: the core sense and two or three uses; 10: partner explanations; 10: choice items; 5: one short paragraph using the form. Record what the paragraph shows and claim no retention untested; Explicit teaching is more effective for EFL copular be learning but its effect is not durable [~M] is why no lasting change should be claimed
The brief names its own routine (its own error codes, rubric, textbook list of uses, writing task) Keep it. Use the brief's categories as the uses to derive from the core sense, keep its rubric and task as given, and add only the diagram, one explanation prompt and one choice check untested
Learners' own writing or work documents are private or sensitive The learner chooses which samples are used and who sees them; they tally their own uses privately and share only examples they choose, with names removed untested
The brief wants a whole performance kept whole (an essay, a presentation, a report for a real audience) Do the concept work before drafting, and give feedback on the form inside the finished product; do not split the performance into gap-fill items untested
Adults in work settings Draw examples and the diagnostic sample from their own genre (emails, reports, client calls); state the form's use in terms of what it does for the reader in that genre untested
Learners with low confidence or low literacy Lean on the picture and on gesture; keep grammatical terms out; let a learner show a choice by pointing or acting before explaining it untested
High stakes, or an assessment with its own item format Practise in the assessment's format, as the brief gives it, and add one sample from the learner's own writing beside it so that test success is not read as use untested

Seven of the eighteen rows cite a claim, and all carry a finding to a setting it did not test (prepositions taught to advanced learners; rule practice; one learner's monitoring; one corpus of argumentative essays); eleven are untested proposals.

Observation, state and explanation

What can be observed is a use of the form in a stated task: the context, whether the form was chosen, avoided or misused, what the learner would have said in the L1, whether the task was timed or edited, and the reason the learner gives when asked. "Has the concept", "is transferring from the L1" and "avoids the form" are inferred states.

Response Plausible explanations Observation that could change the interpretation
Overuses one form (as Thai writers did with because) The L1 equivalent covers more contexts; the learner knows few alternatives; the form was taught early as a safe all-purpose choice Ask what the learner would use in the L1 in those contexts; give a choice task that offers the alternatives and see whether they are known. Learner writers overall overused more connectives than they underused, and drew on a limited range (L2 writers tend to overuse more discourse connectives than they underuse [+W]).
Never errs on the form It is known; it is avoided; the tasks so far never required it An elicitation task in which the form is the natural choice, and a comprehension item (Error analysis has stated limitations [~W]).
Correct on gap-fill items, wrong in own writing The choice is known but not yet used without time to think; the gap-fill cued the answer; the writing genre differs from the practice items Compare a timed with an untimed version of the same item, and a cued with an uncued one (Practice of grammar rules improves only the practiced skill [~M]).
Can explain the core sense but still chooses wrongly The explanation is held but not used under load; the extension from the core sense to this use was not clear; this use is conventional and the core sense does not predict it Ask for the reason at the moment of choice; test a new context the core sense does predict and one it does not.
Chooses correctly but cannot explain Learned from exposure; the explanation needs language the learner lacks Allow the explanation in the L1, or by pointing at the diagram; check a new context.
Is less accurate when editing carefully than when writing freely Conscious checking overrides an intuition that was right; the rule being applied is wrong Compare edited and unedited versions of the same text, form by form (Monitoring lowers accuracy for unlearnable rules [~W], one learner).

These rows are proposals, untested as diagnostics. An elicitation task is also practice, so record each one, and keep open the explanations that remain open.

Evidence and qualifications

No claim in the wiki compares teaching the conceptual motivation of an L2 form with teaching its structure or teaching it as fixed phrases. The claims below bear on parts of the relationship; all but one come from theoretical, review or single-learner sources, and none prints an effect size.

What limits or qualifies the model:

Learners, forms and sources differ across these claims, so do not rank them by label. None measures appropriate use of a form in learners' own unplanned production at a delay.

The evidence by question

Further evidence, not yet read against this model

Objective and learner-valued goal

The designer's objective is usually that learners choose the form where proficient users of the L2 would, in their own writing or speech, at a stated horizon. Elicit separately what the learner wants: to be understood, to sound natural, to stop being corrected, to pass an examination, to write a report a manager will accept. Agreement and divergence are both observations. A learner who values being understood may see no point in a form whose misuse never blocks understanding; one who values sounding natural may care about exactly the subtle differences this page targets.

For example, an engineer writing reports in English may want her reports to read as a native colleague's do, while the course objective for the week is prepositions of time and place in a textbook unit. Teaching the core senses of at, on and in serves the course objective; a re-sample of prepositions in her next real report tests hers. Record both, so that a textbook score is not read as the capability she came for.

What would revise this model?

The expectation should weaken if studies comparing conceptual explanation with structural or phrase-based teaching of the same form, with comparable learners and measures of use in the learners' own production, show no advantage, or show that a gain at post-test is gone at a delay. If explanation helps learners only to judge or edit the form and not to use it in unplanned speech, the model should say so and move its weight to practice. If deviations attributed to the L1 turn out, on checking, to be shared by learners of many L1s, the contrast step should give way to L2-internal contrasts. If the core-sense account does not predict the uses learners get wrong, it is the wrong account for that form. Do not protect the model by calling every failure "not enough practice" after the fact.

Understanding the core sense, choosing correctly on practice items, choosing correctly in one's own writing, and doing so weeks later are separate claims. The present evidence establishes none of the doses or thresholds given above.

Claims

Examples

Key Sources

  • Lb0
  • Sáez, N. (2014). Enhancing the Role of Meaning in the L2 Classroom: A Cognitive Linguistics Perspective. Teachers College, Columbia University Working Papers in TESOL & Applied Linguistics. https://eric.ed.gov/?id=EJ1176947
Page metadata
Field Value
type principle
id teach-conceptual-underpinnings-of-l2-forms
title Teach the conceptual underpinnings of L2 forms rather than only their structural aspects or formulaic units
description For an L2 learner who produces a form but distributes it unlike proficient users because the L1 divides the meaning differently, explaining the meaning that motivates the form's uses (its core sense, with diagrams) and then practising choices across contexts is expected to yield more appropriate use than structural or phrase-only teaching; no claim in the wiki tests this comparison.
status review
canonical True
generated by claude/unspecified on 2026-10-15
trust unverified — no one has confirmed this page's content
sources 1 — lb0

Frontmatter is the page's machine-readable record — see the Schema & Guide.