Level 5 — Advanced · Technical & Conceptual Reading
Lesson 77: Advanced Reading: Fact, Interpretation, Opinion, Evaluation & Inference
The sixth part of this course's argument-analysis arc — separating a text's six reasoning layers (observation, data, interpretation, inference, evaluation, recommendation), and two concrete tests for catching evaluation hidden inside fact-like language.
Learning Objectives
By the end of this lesson, you should be able to:
- Distinguish observation, interpretation, inference, evaluation, opinion, and recommendation as six increasingly interpretive layers, and locate exactly where a given sentence sits among them
- Detect evaluation embedded inside fact-like language, using the 'remove the adjective' and 'replace with neutral language' tests to separate an event from the judgment attached to it
- Distinguish 'shows' from 'suggests,' and treat certainty words like 'obviously,' 'clearly,' and 'undoubtedly' as stance signals rather than evidence
- Identify competing interpretations of the same evidence, and locate an argument's weakest link by comparing its evidence's actual scope against its recommendation's claimed scope
- Tag a paragraph's sentences by their reasoning layer, and identify the author's overall purpose — describing, evaluating, recommending, warning, or predicting
Introduction
Lesson 76 asked what a text shows and hides. Today's lesson — the sixth part of this course's argument-analysis arc — asks a related but distinct question: within what a text does say, which parts are observation, which are interpretation, and which are pure judgment? Without this distinction, a reader easily absorbs an author's interpretation as if it were a plain fact — one of the most common, and most consequential, misreadings in advanced English.
Fact and Interpretation Are Not the Same
Three sentences, three different certainty levels
The automated system reduced processing time by 28% is a measurable claim. The automated system significantly improved operational efficiency leans toward interpretation and evaluation. The company made the right decision by adopting the system is evaluation, close to opinion. All three can appear in the same paragraph — but their certainty, and what evidence would be needed to support each one, is not remotely the same.
Observation
What was directly seen, recorded, or measured
System logs recorded a 28% reduction in average processing time. This is observation — raw data. What that data actually means is a separate, later step.
Interpretation
Explaining what an observation means
Observation: processing time decreased by 28%. Interpretation: the system appears to have improved processing efficiency. The author is now assigning meaning to the raw observation — a genuinely different act from reporting the observation itself.
Inference
Connecting back to Lesson 72's inference bridge
Processing time decreased by 28%, while the workflow remained otherwise unchanged. This suggests that automation contributed to the improvement. Evidence → reasoning → inference: a conclusion built by reasoning from available evidence, not observed directly.
Evaluation
Connecting back to Lesson 65 — a judgment that always needs a criterion
The system is efficient. The approach is practical. The solution is preferable. The implementation was successful. None of these should be accepted as complete meanings on their own. Ask, every time: efficient according to which criterion? Successful according to what measure?
Opinion
Personal judgment or preference — not automatically unsupported
I believe the company should continue using the system. A personal stance, stated explicitly. Opinion is not automatically unsupported — an expert's opinion can be genuinely evidence-informed — but opinion ≠ automatically fact, regardless of how confidently it's stated.
A Six-Level Reading Model
From raw observation to a fully interpretive recommendation
The same underlying event, moving through all six levels
Processing time fell by 28%. → Automation appears to have improved workflow speed. → The system may improve operational efficiency. → The system is a promising solution. → Management should expand its use. Each step becomes progressively more interpretive than the one before it — and each step, in principle, needs its own supporting justification, not a free ride from the step before it.
Why This Matters: The Interpretive Jump
A short evidence claim can carry an enormous conclusion on its back
The pilot produced a 28% reduction in processing time is evidence-oriented. Therefore, the project was an unequivocal success is a very large interpretive jump — success plausibly needs criteria like cost, reliability, accuracy, maintenance, employee acceptance, and scalability, none of which the processing-time figure alone establishes.
"Shows" Versus "Suggests"
Connecting back to Lessons 66, 69, and 71's evidentiary-verb work
The results show that automation improves efficiency makes a strong interpretive claim. If the data actually shows only that processing time decreased, the more honest wording is the results suggest that automation may improve processing efficiency. Ask, every time: does the evidence really "show" this, or merely "suggest" it?
Fact-Like Language Can Hide Evaluation
Connecting back to Lesson 76's loaded language — now inside single sentences
The company wisely invested in automation. "Invested in automation" is the event. "Wisely" is evaluation, smuggled into the same clause. The company irresponsibly ignored employee concerns does the same thing more strongly — "ignored," and especially "irresponsibly," carry real evaluative weight dressed as plain description.
Separate the Event from the Judgment: A Technique
A concrete way to isolate what's actually being claimed
Management recklessly replaced the manual system. Break it apart. Event: management replaced the manual system. Evaluation: recklessly. Now ask: what evidence actually supports "recklessly"? This split — event on one side, judgment on the other — is one of the most immediately useful habits in this entire lesson.
Attribution Changes Meaning
Connecting back to Lesson 75's attribution tracking
Researchers found that automation reduced errors attributes the claim to researchers. The author concludes that automation reduced errors attributes it to the author's own reasoning. Critics argue that automation reduced errors only because workloads temporarily declined introduces a competing interpretation. Always track: who is actually saying what?
"According to" Is a Signal
Connecting back to Lesson 75
According to the company, the new platform improved efficiency. The author is relaying a claim, not necessarily endorsing it. "According to X" means X's claim — not automatically the author's own established fact.
"It Is Clear That…"
A phrase that asserts certainty without supplying evidence
It is clear that automation is the future of logistics. Ask: clear based on what — adoption trends, investment, research, performance data? "Clear" is not itself evidence; it's a claim about how confident the author is.
"Obviously," "Clearly," "Undoubtedly"
Certainty language is not evidence
Obviously, the system is better. Clearly, automation is necessary. Undoubtedly, the policy will succeed. These words are strong certainty and stance signals — and certainty language ≠ evidence. A confident-sounding sentence and a well-supported one are two entirely separate things, exactly as Lesson 74's calibration work established.
Interpretation Can Be Reasonable Without Being Proven
A genuinely important middle ground
Processing time decreased after automation was introduced. Interpretation: automation probably contributed to the improvement — reasonable. Automation definitely caused the improvement — a considerably stronger claim. Plausible interpretation ≠ established conclusion — a reasonable reading is not automatically a proven one.
Competing Interpretations
The same evidence can honestly support more than one reading
Evidence: productivity increased after automation. Interpretation A: automation improved productivity. Interpretation B: additional staffing improved productivity. Interpretation C: workload changed. Interpretation D: management introduced process improvements at the same time. Critical reading means checking whether the author's preferred interpretation is genuinely the only plausible one, or simply the one they chose to present.
The Evidence-to-Interpretation Gap
Connecting back to Lesson 73's measurement gap
Employees reported higher satisfaction after the new system was introduced. Therefore, the system is objectively easier to use. Gap: a satisfaction report → an objective usability claim. Closer to the evidence: employee reports suggest that users perceived the new system as easier to use — the perception is real evidence for a claim about perception, not automatically for a claim about objective usability.
Evaluation Needs Criteria, Revisited
Connecting back to Lessons 65 and 74
The new system is better. Better in terms of speed, cost, accuracy, security, usability, scalability? An evaluation without a stated criterion remains incomplete, no matter how confidently it's phrased.
Recommendation Is Another Step
Each stage needs its own justification, not a free ride from the one before it
The system reduced processing time by 28% — evidence. The system appears to improve operational efficiency — interpretation. The system is a promising investment — evaluation. The company should deploy it across all terminals — recommendation. Evidence requirements generally increase at each successive stage.
The "So What?" Test
Four questions that trace a paragraph from event to action
This traces event → meaning → judgment → action
Running these four questions in order on any dense paragraph reliably surfaces its full reasoning chain, layer by layer.
A Full Worked Example
Observation, data, self-reported evidence, interpretation, evaluation, and recommendation, identified
A port authority introduced automated billing in three terminals. During the first six months, average processing time fell by 28%, while employee-reported billing difficulties declined. Management interpreted these findings as evidence that the system had substantially improved operational efficiency and recommended extending the system to all terminals.
Observation: automated billing was introduced. Data: processing time fell by 28%. Self-reported evidence: employee-reported difficulties declined. Interpretation: findings indicate improved operational efficiency. Evaluation: "substantially improved." Recommendation: extend to all terminals.
Where Is the Weakest Link?
Connecting back to Lesson 73's generalization gap
Which part of the example above needs the most additional support? Very likely the extension to all terminals — the evidence came from only three terminals over six months. Evidence scope < recommendation scope: a generalization gap sitting right at the argument's final, highest-stakes step.
Fact Versus Opinion Is Not Always Binary
Advanced texts move through several intermediate layers, not just two categories
The system reduced processing time (relatively factual) → this suggests improved workflow efficiency (interpretation) → the system is highly effective (evaluation) → it should be adopted across the organization (recommendation) → this is clearly the right strategic choice (strong judgment). Sorting a text into only "fact" and "opinion" misses everything happening in between.
A Tagging System for Reading
Ten short tags for marking a paragraph's reasoning layers directly
A tagged paragraph
[DATA] Processing time decreased by 28%. [INT] This suggests improved workflow efficiency. [EVAL] The system appears highly effective. [REC] Management should expand its use. Tagging even a short passage this way makes its logical layers visible almost instantly.
Why Authors Mix These Layers
Persuasive writing often blends interpretation into fact-shaped sentences
Good writing generally moves evidence → interpretation → conclusion in a traceable sequence. Persuasive writing can present interpretation as though it were fact: the company wisely adopted automation packs event and evaluation into one clause. A critical reader pulls them back apart.
The "Remove the Adjective" Test
A fast, reliable way to expose evaluative language
The company made a remarkably successful investment. Remove the adjective: the company made an investment. Now ask: what evidence establishes "remarkably successful"? This test works on any evaluative word — successful, effective, efficient, irresponsible, innovative, outdated, excellent, superior.
The "Replace With Neutral Language" Test
A second, complementary technique for the same job
The government imposed a harsh regulation. Neutral version: the government introduced a regulation that restricted X. The comparison reveals "harsh" as evaluation, while "restricted X" is a more specific, neutral description. Building a neutral paraphrase is a reliably useful habit in critical reading.
Author Position Detection
What is the author actually doing in this passage?
The same topic, several possible purposes
Two passages about the identical system can have entirely different purposes — one purely descriptive, another building toward a warning or a recommendation. Naming the purpose explicitly clarifies what kind of claim the passage is actually making.
Lesson 77 Master Framework
The full reasoning-layer architecture, integrated with this arc's earlier tools
Vocabulary in Context
interpretive jumpnoun phrasecourse term
a step in reasoning that moves from evidence to a considerably stronger or broader claim than the evidence alone establishes (ব্যাখ্যামূলক লাফ)
Often with make an interpretive jump, a large interpretive jump, interpretive jump from
“Concluding that morale collapsed from one resignation is an interpretive jump.”
attributed claimnoun phraseformal
a statement credited to its source ('researchers found', 'according to the company') rather than stated as the author's own fact (আরোপিত দাবি)
Often with an attributed claim, treat as an attributed claim, attributed claim such as
“'According to the company, costs fell' is an attributed claim, not a proven fact.”
evaluative adjectivenoun phrasetechnical
a descriptive word that carries a judgment of quality or worth rather than neutral description, such as 'wisely' or 'reckless' (মূল্যায়নমূলক বিশেষণ)
Often with use an evaluative adjective, loaded evaluative adjective, evaluative adjective such as
“'Reckless' is an evaluative adjective; 'unapproved' would be a neutral description.”
neutral paraphrasenoun phrasetechnical
a restatement of a sentence with its evaluative or loaded language replaced by plain, descriptive wording (নিরপেক্ষ পুনর্বিবৃতি)
Often with write a neutral paraphrase, provide a neutral paraphrase, neutral paraphrase of
“A neutral paraphrase of 'the reckless plan' would be 'the plan that lacked safety approval'.”
authorial purposenoun phraseformal
what an author is actually doing in a passage — describing, evaluating, warning, recommending, or predicting — as distinct from the topic itself (লেখকের উদ্দেশ্য)
Often with identify the authorial purpose, clear authorial purpose, authorial purpose behind
“The authorial purpose of the memo is to warn managers, not to describe last year's results.”
inferencenounformal
a conclusion reached by reasoning from available evidence, rather than observed directly (অনুসিদ্ধান্ত)
Often with draw an inference, reasonable inference, inference from
“A reasonable inference from the gate logs is that most trucks arrived outside their slots.”
recommendationnounformal
the final, most action-oriented stage of an argument, urging a specific decision, and typically requiring the most supporting evidence of any stage (সুপারিশ)
Often with make a recommendation, formal recommendation, recommendation to
“The audit's main recommendation is that the port introduce independent safety inspections.”
evidence-to-interpretation gapnoun phrasecourse term
the distance between what evidence directly shows and a broader interpretive claim built on top of it (প্রমাণ-থেকে-ব্যাখ্যার ব্যবধান)
Often with an evidence-to-interpretation gap, spot the evidence-to-interpretation gap, evidence-to-interpretation gap between
“Data showing fewer accidents leaves an evidence-to-interpretation gap if the report credits the new policy without proof.”
Guided Reading Practice
Read this passage once, straight through. Then tag each sentence using this lesson's system (OBS, DATA, INT, EVAL, REC, and so on), apply the "remove the adjective" test to any evaluative language, and identify the weakest link, before checking your reading against the notes that follow.
A regional airline introduced a new baggage-tracking app across five hub airports. Over three months, lost-baggage reports fell by 22%, and the operations team wisely credited the app's real-time scanning feature. This clearly demonstrates that the app dramatically improved baggage handling, and the airline should roll it out network-wide immediately.
Tags: [DATA] "lost-baggage reports fell by 22%" — measured evidence. [ATTR + EVAL] "the operations team wisely credited the app's real-time scanning feature" — an attributed claim, with "wisely" as an embedded evaluation of the team's own judgment, not evidence about the app itself. [INT + EVAL] "this clearly demonstrates that the app dramatically improved baggage handling" — "clearly demonstrates" and "dramatically" are certainty and evaluative language layered onto a single data point. [REC] "the airline should roll it out network-wide immediately" — recommendation. Remove-the-adjective test: "dramatically improved" → "improved" — what specifically established the size of the improvement beyond the 22% figure already given? Weakest link: evidence comes from five hub airports over three months; the recommendation claims the entire network, immediately — a generalization and time gap sitting directly underneath the passage's most confident language.
Golden Rule
Golden Rule
Never treat an author's interpretation, evaluation, or recommendation as if it were raw evidence — the most important question for any sentence is: what is observed, what is inferred, and what is merely judged?
Lesson Summary
This lesson gave you a six-level model — observation, data, interpretation, inference, evaluation, recommendation — for locating exactly how interpretive any given sentence actually is. You practised detecting evaluation hidden inside fact-shaped language using two concrete techniques: removing the adjective, and building a neutral paraphrase. You separated "shows" from "suggests," treated certainty words like "obviously" and "clearly" as stance rather than evidence, and held "plausible interpretation" and "established conclusion" apart as two genuinely different claims. You identified competing interpretations of the same evidence, tagged a full paragraph using a ten-symbol system, and closed by naming an author's overall purpose — describing, evaluating, recommending, warning, predicting — rather than assuming every passage on a topic is doing the same job.
Six lessons, one connected discipline
Lessons 72 through 77 have built one continuous reading capability: an argument's anatomy, its missing bridges, its evidence sufficiency, its evidence's actual source and quality, what it shows and hides, and now, exactly how interpretive each of its sentences actually is. Together, they replace a single vague reaction — "this seems convincing" or "this seems weak" — with a precise account of which specific claim, at which specific layer, is or isn't earning the confidence it's stated with.
Practice: Test What You've Learned
Work through every question yourself before checking anything.
Before you start
For every sentence below, tag its reasoning layer using this lesson's system (OBS, DATA, INT, INF, EVAL, OP, REC, ATTR, or QUAL) — "this sounds like a claim" is not yet a complete answer.
Part A — The Six Levels
- 'Average response time dropped from 12 minutes to 7 minutes.' Tag this sentence's reasoning layer.
Show answer
DATA — measured evidence. It reports a directly recorded numeric change with no interpretation, evaluation, or recommendation attached yet. - 'This drop suggests the new ticketing system helped agents resolve issues faster.' Tag this sentence's reasoning layer, and explain what distinguishes it from question 1.
Show answer
INT — interpretation, carried by the phrase "this suggests." Unlike question 1, which simply reports the measured change, this sentence assigns a possible cause and meaning to that change, an interpretive step layered on top of the raw data. - 'The company should adopt the ticketing system across every department.' Tag this sentence's reasoning layer, and identify what additional evidence this layer typically requires.
Show answer
REC — recommendation, the most action-oriented and interpretive layer. It typically requires evidence beyond a single benefit, such as data confirming the improvement holds across departments with different needs, along with cost, reliability, and implementation considerations.
Part B — Detecting Hidden Evaluation
- 'The manager boldly overhauled the entire approval process.' Apply the "remove the adjective" test, and state what evidence would be needed to justify the removed word.
Show answer
Removing the adjective leaves: the manager overhauled the entire approval process, the plain event. Justifying "boldly" would require evidence about the risk involved, the opposition faced, or the courage required in making that decision, none of which the bare event alone establishes. - 'The startup foolishly ignored early warning signs from customers.' Separate the event from the judgment, using this lesson's technique.
Show answer
Event: the startup did not act on early warning signs from customers. Judgment: "foolishly," an evaluative label characterizing that inaction as unwise, a separate claim that would need its own justification, such as evidence that acting on the signs would clearly have prevented harm. - Rewrite 'the committee stubbornly refused to reconsider the policy' as a neutral paraphrase.
Show answer
Neutral paraphrase: the committee did not reconsider the policy after being asked to. This removes "stubbornly," which frames the decision as an unreasonable character trait rather than simply reporting what happened.
Part C — Shows vs. Suggests, and Certainty Language
- 'The data show that remote work always increases productivity.' Explain why "show" combined with "always" makes this a particularly strong claim, and what would justify toning it down.
Show answer
"Show" claims the data directly demonstrates the conclusion, and "always" claims a universal, exceptionless pattern, together making an extremely strong and easily falsifiable claim. Toning it down would be justified if the data only reflects some contexts or a general tendency, in which case wording such as "suggest that remote work often increases productivity" would more honestly match the evidence. - 'Clearly, this is the best option available.' Explain what is missing from this sentence despite its confident tone.
Show answer
Missing: any stated criterion for "best," such as cost, speed, or quality, and any actual evidence supporting the comparison to the other options. "Clearly" signals the author's certainty, not proof, so the sentence sounds confident while supplying no verifiable support for its claim. - Explain, in your own words, why 'plausible interpretation' and 'established conclusion' are not the same claim.
Show answer
A plausible interpretation is merely a reasonable reading consistent with the evidence, possibly one of several such readings. An established conclusion has been confirmed with enough evidence to rule out competing explanations, so treating a plausible interpretation as an established conclusion skips the additional verification that closing that gap actually requires.
Part D — Competing Interpretations and Weakest Links
- Sales increased the same month a new pricing model launched and a major competitor raised its prices. A report credits the pricing model alone. Identify at least one competing interpretation the report ignores.
Show answer
A competing interpretation is that the competitor's price increase, not the new pricing model, drew customers toward this company and drove the sales increase, a plausible alternative explanation the report ignores entirely. - A three-week trial in one department found improved morale after a new benefits policy. The company announces the policy will be rolled out company-wide. Identify the weakest link in this argument, using this lesson's terms.
Show answer
The weakest link is the recommendation stage: the evidence's scope, one department over three weeks, is far narrower than the recommendation's scope, the whole company, a clear case of evidence scope falling short of recommendation scope. - Explain why tagging a paragraph's sentences by reasoning layer is more useful than simply labeling the whole paragraph 'persuasive' or 'objective.'
Show answer
A single paragraph often blends several reasoning layers, observation, interpretation, evaluation, and recommendation, each with its own evidentiary requirement, so one blanket label hides exactly where the reasoning shifts from solid data to looser judgment. Tagging sentence by sentence lets a reader pinpoint precisely which specific claim is well-supported and which is not, rather than accepting or rejecting the whole paragraph at once.
Part E — Full Passage Analysis
Return to this lesson's guided reading passage about the baggage- tracking app.
- Tag each sentence in the passage using this lesson's system.
Show answer
Sentence 1, "a regional airline introduced a new baggage-tracking app...": OBS/background. Sentence 2, "lost-baggage reports fell by 22%, and the operations team wisely credited...": DATA plus ATTR with an embedded EVAL in "wisely." Sentence 3, "this clearly demonstrates... and the airline should roll it out...": INT plus EVAL in the first clause, then REC in the recommendation clause. - Identify the evaluative language the passage uses, and apply the "remove the adjective" test to it.
Show answer
The evaluative language is "wisely" in "wisely credited" and "dramatically" in "dramatically improved," plus the certainty phrase "clearly demonstrates." Removing the adjectives leaves "the operations team credited the app's real-time scanning feature" and "this demonstrates that the app improved baggage handling," plain claims that still need their own justification beyond the 22% figure already given. - Identify the passage's weakest link, and explain why it sits at the recommendation stage.
Show answer
The weakest link is the recommendation to roll the app out network-wide immediately. The evidence comes from only five hub airports over three months, so both the recommendation's scope, the entire network, and its urgency, immediately, exceed what that narrower, shorter-term evidence actually supports. - Explain why summarising this passage as "the app fixed the airline's baggage problem" would overstate what the passage's own evidence establishes.
Show answer
That summary treats a 22% reduction across five hub airports over three months as a complete, general solution to baggage handling, when the passage's own evidence supports only a partial, limited improvement, with the causal credit not even fully isolated from other factors like the operations team's own actions. It converts a data-plus-interpretation claim into an unqualified success claim the evidence never actually established.
Next: Lesson 78 — Advanced Reading: Author's Purpose, Rhetorical Moves & Argument Function: why understanding a sentence's meaning is not the same as understanding why the author put it there.