How to Read | Class No.0016 | The Uncertainty Read — Hold Ambiguity, Probability, Missing Evidence and Competing Meanings Without Guessing

Class No.0003: The Evidence Read · Class No.0004: The Critical Read · Class No.0012: The Quantitative Read

Uncertainty is not the same as confusion

Confusion means the reader’s model is incomplete or unstable.

Uncertainty can remain even after excellent comprehension.

A well-designed mystery may leave motive uncertain. A scientific result may have a confidence interval. A policy forecast may depend on future assumptions. A historical source may support two plausible interpretations.

Do not repair genuine uncertainty as if it were a reading error.

The first question: is the uncertainty in me or in the text?

Ask:

• Did I miss a word?
• Did I lose the referent?
• Did I overlook a condition?
• Does the text actually provide insufficient information?
• Are multiple interpretations deliberately possible?

The repair depends on the source of uncertainty.

Resolve reader uncertainty before declaring textual ambiguity

Before saying “this could mean anything,” check:

1. literal meaning;
2. syntax;
3. reference;
4. context;
5. later evidence.

Many apparent ambiguities disappear after careful reading.

Ambiguity means more than one viable reading

Consider:

Ana saw the teacher with the telescope.

Who has the telescope?

The grammar permits more than one reading without context.

A strong reader does not choose arbitrarily. They wait for evidence.

Lexical ambiguity

A word can have multiple senses.

Bank may refer to a financial institution or river edge.

Light may refer to illumination, weight or seriousness.

Context usually resolves the sense.

If it does not, preserve both until more evidence appears.

Syntactic ambiguity

Sentence structure can permit multiple attachments.

Example:

She discussed the problem with the manager in the office.

Was the discussion in the office, or was the manager in the office?

Most real contexts resolve this. If not, the sentence is genuinely ambiguous.

Referential ambiguity

Pronouns can point to more than one plausible noun.

Maya called Lina after she finished work.

Who finished work?

A reader should not silently decide and then forget that the decision was inferential.

Pragmatic ambiguity

A sentence may mean more than its literal proposition.

“It’s getting late.”

It may be a statement, a suggestion to leave, a warning or an excuse.

Speaker relationship and situation matter.

Literary ambiguity can be productive

A poem or story may deliberately resist one final interpretation.

The reader’s job becomes comparative:

• What supports Reading A?
• What supports Reading B?
• Which details each reading explains?
• What remains unresolved?

Ambiguity can create depth rather than error.

Competing interpretations need common criteria

Do not compare interpretations by which sounds more interesting.

Compare:

• textual coverage;
• evidence strength;
• number of unsupported assumptions;
• ability to explain counterevidence;
• fit with structure and context.

The stronger interpretation usually explains more with less forcing.

Probability language expresses uncertainty

Words such as:

possible, plausible, likely, probable, unlikely, rare, may, might, could, tends to, usually, often, sometimes

do not all mean the same thing.

Readers should preserve these levels rather than flattening them into certainty.

Modal verbs control claims

Compare:

• will;
• should;
• may;
• might;
• could.

A headline that turns “may increase risk” into “causes harm” has strengthened the claim.

The Uncertainty Read tracks modality.

Hedging is not always weakness

Academic and scientific writers often hedge because evidence has limits.

Phrases such as:

• the results suggest;
• within this sample;
• under these conditions;
• may be associated with;
• one possible explanation;

can signal methodological discipline.

Do not mistake calibrated language for lack of knowledge.

Certainty should track evidence

Class No.0003 established this principle.

Use a confidence scale:

• High: explicit or converging evidence; alternatives weak.
• Moderate: best-supported interpretation but plausible alternatives remain.
• Low: possible but underdetermined.
• Unknown: evidence insufficient to rank options responsibly.

Confidence is part of the answer.

“Unknown” is different from “50–50”

If evidence is missing, assigning equal probability may be unjustified.

“I do not know” does not mean both outcomes are equally likely.

It means the current information does not support a precise probability.

Missing evidence has types

Ask what is missing:

• fact;
• sample;
• comparison group;
• timeline;
• definition;
• source attribution;
• method;
• counterfactual;
• independent corroboration.

“More evidence” is too vague.

The Missing Evidence question

Ask:

What single additional piece of information would change my confidence most?

This converts uncertainty into an information need.

Unknown cause

An event can be known while its cause remains uncertain.

Do not turn sequence into causation merely to remove discomfort.

Say:

X occurred before Y, but the text does not establish that X caused Y.

Unknown motive

A character or historical actor may act without clear motive.

Possible motives can be ranked by evidence, but some uncertainty may remain.

Do not convert interpretation into biography.

Unknown future

Forecasts are conditional.

Read:

• scenario;
• assumptions;
• range;
• time horizon;
• confidence.

A projection is not a promise.

Uncertainty in quantitative texts

Class No.0012 introduced margins of error, confidence intervals and measurement uncertainty.

The Uncertainty Read adds the linguistic layer:

How does the writer translate numerical uncertainty into words?

Does the headline preserve the interval or present a point estimate as fact?

Ranges are information

A forecast of 3–5% growth is not an annoying refusal to give one number.

The range expresses uncertainty.

Ask what assumptions could move the outcome within or beyond the range.

Point estimates can create false precision

A model may estimate 2.7.

That does not always mean reality is known to one decimal place.

Look for uncertainty around the estimate.

Conflicting sources are not automatically equal

Two sources disagreeing does not create a 50–50 situation.

Compare:

• expertise;
• method;
• evidence;
• recency;
• independence;
• scope.

Conflict should be evaluated, not averaged.

OECD treats conflict handling as advanced reading

The PISA 2018 reading framework explicitly includes detecting and handling conflict within and across texts among complex reading processes. It notes that multiple texts may provide inconsistent information and readers need evaluation processes to acknowledge and handle that conflict.

Mature reading therefore includes unresolved disagreement.

First identify whether sources truly conflict

Apparent conflict may disappear when you align:

• definitions;
• time periods;
• populations;
• measures;
• conditions.

Source A may say unemployment rose in June; Source B may say annual unemployment fell. Both can be true.

Different questions produce different answers

A study may ask whether a policy raises average achievement.

Another may ask whether it narrows inequality.

Different outcomes are not necessarily contradictory.

Align the question before comparing conclusions.

Value disagreement is not factual uncertainty

Two writers can agree on facts and disagree about what should be done.

That is a normative conflict.

More empirical data may help, but it may not eliminate the underlying value trade-off.

Ambiguous task versus ambiguous text

An examination or assignment question can itself be unclear.

Before blaming the passage, restate the task:

• What answer form is required?
• Which term is ambiguous?
• What reasonable interpretation best fits context?

SETC’s exam-specific ambiguity pages own timed response handling. This class owns the general reading behaviour.

Uncertainty and curiosity

Uncertainty can be productive because it identifies where inquiry should go next.

Ask:

• What do I know?
• What do I suspect?
• What remains unknown?
• What evidence would discriminate among possibilities?

This is the bridge from reading to research.

Uncertainty and anxiety

Some learners experience uncertainty as a demand to answer faster.

Train a different response:

I have two plausible readings. I will keep both until I find a deciding clue.

This is controlled delay, not indecision.

Premature closure

Premature closure occurs when the reader picks the first plausible interpretation and stops looking.

It is common when:

• time pressure is high;
• the reader strongly prefers one answer;
• ambiguity feels uncomfortable;
• one clue is vivid.

Repair: generate one alternative before finalising.

Perpetual doubt is the opposite failure

Some readers refuse to commit even when evidence is strong.

Critical thinking does not require endless scepticism.

When one interpretation clearly dominates, confidence should increase.

Calibration means moving both directions

A calibrated reader can become:

• less certain when evidence weakens;
• more certain when evidence converges.

Both movements are rational.

The Evidence–Confidence pair

For each conclusion, write:

Claim: ______
Confidence: high / moderate / low
Reason: ______

This makes uncertainty explicit.

The Competing Meanings table

Interpretation	Supporting evidence	Counterevidence	Confidence
A	…	…	…
B	…	…	…

Use this only when ambiguity is real enough to justify the effort.

The Unknowns ledger

For research reading, maintain:

• known;
• probable;
• contested;
• unknown;
• evidence needed.

This prevents the final synthesis from presenting all information at the same certainty level.

Kindergarten: two possible explanations

Use a picture book scene.

Ask:

Why might the character be hiding?

Accept two plausible ideas, then ask what later picture or sentence could help decide.

Children learn that not every question has an immediate answer.

Primary 1: “we don’t know yet”

Teach this phrase explicitly.

If a story has not revealed who knocked at the door, do not guess as fact.

Say:

We don’t know yet. What clues do we have?

Primary 2: possible versus certain

Use two labels:

text tells us / maybe

Build the distinction before introducing more formal confidence language.

Primary 3: alternative inference

Give one ambiguous character action.

Ask for two explanations and one clue that would favour each.

Primary 4: probability words

Compare:

• will;
• probably;
• might;
• unlikely.

Ask learners to order certainty and explain why wording matters.

Primary 5: missing information

Give an incomplete notice or short report.

Ask what information is needed before making a decision.

Uncertainty becomes a search target.

Primary 6: conflicting simple sources

Use two age-appropriate accounts of the same event.

Ask:

• what both agree on;
• where they differ;
• what would help resolve the difference.

Secondary 1: ambiguous pronouns and sentences

Give syntactically ambiguous examples.

Students identify possible readings and rewrite the sentence to remove ambiguity.

Reading uncertainty improves writing clarity.

Secondary 2: modality in nonfiction

Highlight:

may, might, likely, appears, suggests, tends to.

Ask what would change if each were replaced with proves or always.

Secondary 3: source conflict

Use two credible texts that disagree.

First align definitions and scope.

Then compare evidence.

Students write a conclusion that preserves the unresolved part.

Secondary 4: uncertainty under time limits

Teach a decision threshold:

• if evidence strongly favours one interpretation, proceed;
• if two remain close and task allows, qualify;
• if the question demands a choice, select the best-supported and note the deciding clue mentally.

Exam execution belongs to SETC’s dedicated performance owners; the underlying reading skill is calibration.

JC: probabilistic argument

Advanced texts often make claims such as:

Policy X is likely to produce Y under conditions A and B, although outcome Z remains possible.

Do not simplify this into “X causes Y.”

Preserve conditional structure.

JC: incomplete evidence

Ask:

• What does the available evidence support?
• What would require stronger evidence?
• What evidence is inaccessible?
• What remains a value judgement rather than an empirical question?

Mature reading separates the solvable from the unresolved.

JC: multiple-source uncertainty

Build a synthesis with confidence labels:

• well established;
• supported but contested;
• plausible;
• insufficient evidence.

This is more useful than pretending every paragraph has equal certainty.

Uncertainty in literature

Ask whether ambiguity is:

• temporary;
• resolved later;
• structural;
• thematic;
• caused by unreliable narration.

The type of uncertainty changes the interpretation.

Uncertainty in science

Read:

• measurement error;
• sampling uncertainty;
• model uncertainty;
• confidence interval;
• alternative explanation;
• replication status.

Scientific uncertainty is often quantified or methodologically described.

Uncertainty in history

Historical evidence can be incomplete.

Documents may be lost. Witnesses may disagree. Motives may be inferred.

A strong historian distinguishes:

• established event;
• probable interpretation;
• contested claim;
• unknown.

Uncertainty in news

Breaking news evolves.

Early reports can be incomplete or wrong.

Look for:

• what is confirmed;
• what officials or witnesses claim;
• what remains unverified;
• when the article was updated.

Recency and uncertainty interact.

Uncertainty in policy

Policy choices often involve forecasts, trade-offs and values.

A reader should distinguish:

• known current conditions;
• modelled future effects;
• uncertain behavioural responses;
• normative preferences.

One category should not masquerade as another.

The Uncertainty Read protocol

1. Establish what is explicit.
2. Resolve reader-side confusion.
3. List viable interpretations.
4. Compare supporting and opposing evidence.
5. Preserve modal and probability language.
6. Identify missing evidence.
7. Align conflicting sources by scope and definition.
8. Assign confidence proportionate to evidence.
9. State what remains unknown.

The Uncertainty Read audit

Ask:

• Am I confused, or is the text genuinely uncertain?
• What is definitely stated?
• What is inferred?
• What alternative readings remain viable?
• Which interpretation explains more evidence?
• What wording signals probability or limitation?
• What evidence is missing?
• Do conflicting sources truly address the same question?
• What confidence level is justified?
• What would change my mind?

Common failure: guess to remove discomfort

Repair: label uncertainty and wait for evidence.

Common failure: every ambiguity becomes “anything goes”

Repair: compare readings by evidence and explanatory power.

Common failure: hedging is treated as weakness

Repair: preserve the author’s calibrated certainty.

Common failure: disagreement becomes 50–50

Repair: compare source strength rather than count opinions.

Common failure: permanent scepticism

Repair: allow converging evidence to increase confidence.

Three learners, three uncertainty failures

Maren chooses the first plausible interpretation because she dislikes ambiguity. Her repair is alternative generation.

Iona notices every caveat and becomes unwilling to conclude anything. Her repair is confidence calibration upward when evidence converges.

Leonie understands the evidence but strengthens “may” into “will” in her summary. Her repair is modal fidelity.

A 45-minute Uncertainty Read tutorial

1. 10 minutes: distinguish ambiguity from confusion.
2. 10 minutes: compare two interpretations.
3. 5 minutes: rank modal language by certainty.
4. 10 minutes: compare two conflicting sources.
5. 5 minutes: identify missing evidence.
6. 5 minutes: write a calibrated conclusion.

Research anchors

• OECD — PISA 2018 Reading Framework
• Education Endowment Foundation — Reading Comprehension Strategies
• Institute of Education Sciences — Disciplinary Literacy Processes and Procedures

Continue through SETC

• How to Read | Class No.0003 | The Evidence Read
• How to Read | Class No.0004 | The Critical Read
• How to Read | Class No.0012 | The Quantitative Read
• English Ambiguity Resolution Performance

Final lesson

A reader does not become sophisticated by producing certainty faster.

A sophisticated reader knows what the evidence settles, what it only suggests and what remains open.

Resolve confusion. Preserve genuine ambiguity. Keep probability language intact. Compare alternatives. Name missing evidence. Increase confidence when evidence converges and reduce it when the case weakens.

Sometimes the most accurate sentence a reader can write is:

This is the best-supported interpretation so far, but the evidence does not yet close the question.

That is not weak reading.

It is calibrated reading.

Return to the English Learning Library · Primary English · Secondary English · Advanced English / General Paper

Explore the English Reading and Comprehension guides