TOEIC Link Grammar — Quantifiers and Determiners: The Countable-Uncountable Discrimination That Separates Band 18 from Band 24 on Reading and Writing
Quantifier and determiner errors are the most frequent low-visibility deductions in the TOEIC Link incomplete-sentence reading questions and the writing-module rubric. Internal corpus analysis shows that band-18 to band-22 candidates make a quantifier or determiner error in roughly one out of every six sentences they write, and that the reading-module incomplete-sentence section contains an average of three quantifier-discrimination items per test form. The error class is low-visibility because each individual mistake costs only a small fraction of a band point, but the errors compound across a response and across a test form, and the cumulative cost is large enough to gate the band-18-to-band-24 transition for candidates who never install the discrimination explicitly.
This guide formalizes the countable–uncountable discrimination, distributes the high-frequency quantifiers across the discrimination, names the article–quantifier interaction rules, and outlines a four-week drill routine. For adjacent grammar topics, see the grammar articles a/an/the guide and the grammar subject-verb agreement guide.
The countable–uncountable discrimination is the foundation
Every quantifier in English carries a countability tag. Some quantifiers select countable nouns only (many, few, several, a number of), some select uncountable nouns only (much, little, a great deal of, an amount of), and some select both (some, any, a lot of, plenty of, most, all, no). A candidate who has not installed the countability tag of each high-frequency quantifier cannot select correctly on the reading-module incomplete-sentence questions and cannot write the writing-module response without producing the high-frequency errors much problems, many information, few advice, little employees.
The countability of a noun is itself a learned fact that does not always transfer from the Japanese mental model. Nouns that are countable in English but uncountable in the Japanese mental model include suggestion, request, opportunity, idea, result, problem, decision, recommendation, criticism, comment, question. Nouns that are uncountable in English but feel countable in the Japanese mental model include information, advice, equipment, furniture, luggage, research, knowledge, evidence, feedback, machinery, stationery, baggage, progress, traffic, vocabulary. The band-18-to-band-24 candidate must memorize both lists and must be able to recall the countability tag of each noun under reading-module time pressure.
The discrimination test
The diagnostic test for any noun is the pluralization–article check: does the noun take a plural -s form, and can the noun take the indefinite article a/an? If both answers are yes, the noun is countable. If neither answer is yes, the noun is uncountable. Mixed answers indicate a polysemous noun with separate countable and uncountable senses (for example, experience is uncountable when it means knowledge gained over time but countable when it means a single event).
The high-frequency quantifier distribution
The eight high-frequency quantifiers below account for roughly 85% of all quantifier-discrimination items on the TOEIC Link reading module and roughly 90% of all quantifier deployments on band-18 to band-24 writing-module responses.
Many — countable only
Many selects countable plural nouns and combines with of to take a partitive structure (many of the candidates). The high-frequency error is many information, which violates the countability tag of information. The corrected form uses much information or a lot of information.
Much — uncountable only
Much selects uncountable nouns and is most natural in negative and interrogative contexts (we do not have much time, is there much demand). In affirmative contexts, a lot of is more natural than much, and the candidate who uses much in affirmative contexts (we have much time) sounds slightly stilted even when the grammar is technically correct. The writing-module rubric does not deduct for the stilted affirmative use, but the candidate who can switch to a lot of in affirmative contexts signals upper-band naturalness.
Few / a few — countable only
Few carries a negative implication (close to none), and a few carries a positive implication (some, but not many). The discrimination is rubric-relevant on the writing module because the candidate who writes few employees attended the meeting communicates that attendance was disappointing, where a few employees attended communicates that attendance was modest but acceptable. The reading-module incomplete-sentence section tests the discrimination by manipulating the surrounding context to make one of the two readings clearly correct.
Little / a little — uncountable only
Little and a little parallel the few / a few discrimination but select uncountable nouns. Little progress carries the negative implication (close to none), and a little progress carries the positive implication (some, but not much). The writing-module high-frequency error is little employees, which violates the countability tag, and few advice, which also violates the countability tag.
Several — countable only
Several selects countable plural nouns and means more than two but not many. The quantifier is unambiguous on countability but is occasionally misused in scope, where the candidate uses several to mean many (several years of research used to mean two decades of research). The reading-module incomplete-sentence section occasionally tests the scope by providing context that distinguishes several (small number) from many (large number).
Some / any — both
Some and any select both countable and uncountable nouns. The discrimination between the two is governed by the affirmative–interrogative–conditional split: some is the default for affirmative contexts, any is the default for negative, interrogative, and conditional contexts. Exception: some appears in interrogative contexts when the speaker expects a positive answer (would you like some coffee?), and any appears in affirmative contexts when the meaning is any whatsoever (any candidate can apply).
A lot of / lots of — both
A lot of and lots of select both countable and uncountable nouns and are the safest fall-back quantifiers when the candidate is uncertain about the countability tag of the following noun. The writing-module rubric does not penalize the informal lots of in body paragraphs but slightly prefers a lot of in formal registers. Plenty of is a near-synonym that adds the implication of sufficient supply.
No / none — both
No selects both countable and uncountable nouns as a determiner (no problems, no information), and none is the corresponding pronoun (none of the candidates, none of the information). The high-frequency error is the doubled negative no + negative verb (he does not have no time), which is non-standard in formal English and is graded as a rubric error on the writing module.
Article–quantifier interaction rules
Quantifiers interact with the definite and indefinite articles in three rule-bound ways that the candidate must control.
Rule 1 — Quantifier + bare noun for general reference
Most quantifiers combine with a bare noun (no article) for general reference: many candidates, much information, few opportunities, little progress. The bare-noun form is the default for general statements.
Rule 2 — Quantifier + of + the/this/these/my/our for specific reference
When the reference is specific, the quantifier takes the partitive of structure with a determiner: many of the candidates, much of the information, few of these opportunities, little of our progress. The partitive structure changes the meaning from general to specific, and the candidate who omits the of the in the specific reading produces a meaning error.
Rule 3 — All / both / half can take the bare partitive
The quantifiers all, both, and half can take a bare partitive without the of in informal registers (all the candidates, both employees, half the budget), but the writing-module rubric prefers the explicit of form (all of the candidates, both of the employees, half of the budget) in formal contexts.
The four-week drill routine
Week 1 — Countability tagging drill
The candidate works through a 200-noun corpus and tags each noun as countable, uncountable, or both. The week's output is a tagged corpus and a self-test that confirms 95% accuracy on the high-frequency countable–uncountable boundary cases (information, advice, equipment, furniture, research, evidence, feedback, progress).
Week 2 — Quantifier-selection drill
The candidate works through 80 reading-module incomplete-sentence items focused on quantifier selection, with detailed review of the eight high-frequency quantifiers above. The week's output is an item-level error log that confirms above 85% accuracy on the quantifier discrimination.
Week 3 — Quantifier-production drill
The candidate writes 30 short paragraphs (roughly 100 words each) on prescribed topics with a deliberate quantifier diversity quota: each paragraph must use at least four different quantifiers from the high-frequency list. The week's output is a per-paragraph quantifier-diversity log and a self-corrected error log.
Week 4 — Integrated writing drill
The candidate writes 12 full TOEIC Link writing-module responses with quantifier-and-determiner discipline as the primary self-grade axis. The week's output is a per-response error count; target: zero countability violations and zero article–quantifier interaction errors across the final three responses.
CEFR band-by-band targets
- Band 18: Quantifier errors in roughly one out of every six sentences; high-frequency countability violations on information, advice, equipment.
- Band 21: Quantifier errors in roughly one out of every twelve sentences; high-frequency violations rare but still occasional; few / a few and little / a little discrimination not yet productive.
- Band 24: Quantifier errors in roughly one out of every twenty-five sentences; countability violations effectively eliminated; few / a few and little / a little discrimination productive.
- Band 27: Quantifier errors effectively zero; partitive of structure deployed correctly in specific reference; quantifier diversity used deliberately for register control.
Closing note
Quantifiers and determiners are a small, finite grammar topic that delivers disproportionate band-point return because the high-frequency errors are common, the discriminations are learnable in four weeks, and the rubric treats consistent quantifier control as a marker of upper-band naturalness. The candidate who installs the eight high-frequency quantifiers with their countability tags and the three article–quantifier interaction rules converts a steady source of low-visibility deductions into a steady source of upper-band signaling within one drill cycle, and the transfer to the speaking module's structured-response sections is automatic because the same discriminations govern correct quantifier selection in extemporaneous speech.