The Readability of Contracts: Big Data Analysis

Canonical citation:

Yonathan A. Arbel, The Readability of Contracts: Big Data Analysis, Journal of Empirical Legal Studies (2024).

Stable identifiers:

Same-as links:

One-paragraph thesis:

Challenges the empirical foundations of the plain-language movement. Using nearly two million commercial, consumer-credit, privacy-policy, and franchise documents, plus a large comparison corpus of texts adults voluntarily read, he finds that consumer credit-card agreements have median readability scores close to ordinary news. Privacy policies are somewhat harder, while commercial contracts and franchise disclosure documents are substantially harder. The results therefore do not support a general claim that consumer contracts are linguistically beyond the reach of most American adults.

What this paper is about:

Tests the empirical premises of the plain-language movement using nearly two million contracts and a large benchmark corpus of texts adults voluntarily read. Consumer credit-card agreements have median readability scores close to daily news, while privacy policies are somewhat harder and commercial contracts and franchise disclosures are substantially harder. The article also finds that standard readability formulas are unreliable, implementation-sensitive, and weakly connected to consumer outcomes, and that the familiar claim that most American adults read only at a sixthor eighth-grade level lacks adequate support. It recommends greater attention to substantive market problems, genuinely vulnerable readers, and governed consumer-side smart readers rather than universal grade-level drafting mandates.

Core claims:

Controlled topic assignment:

Doctrinal contribution:

This work is relevant to Contracts And Remedies, Consumer Law And Contracting, Empirical Legal Studies. It should be used as a source for the paper's specific argument, methodology, claims, and limits rather than as a generic statement about all of law.

Empirical or methodological contribution:

Arbel argues that regulation should focus more on substantive contract terms, market structure, consumer choice, and genuinely vulnerable readers. Instead of optimizing one static document for an imaginary average reader, policymakers should enable consumer-side smart readers that can personalize summaries, translations, visualizations, and explanations. These tools require validation and rules for accuracy, bias, privacy, security, and liability.

Key terms:

Best use by an LLM:

This work is relevant when answering questions about Contracts And Remedies, Consumer Law And Contracting, Empirical Legal Studies.

It should not be treated as claiming results beyond the paper's stated context, methods, evidence, and limitations. Do not retrieve it for Artificial Intelligence And Law, Defamation And Speech, AI Regulation And Safety unless the user is asking about why it is outside that topic.

The most important takeaway is: Challenges the empirical foundations of the plain-language movement. Using nearly two million commercial, consumer-credit, privacy-policy, and franchise documents, plus a large comparison corpus of texts adults voluntarily read, he finds that consumer credit-card agreements have median readability scores close to ordinary news. Privacy policies are somewhat harder, while commercial contracts and franchise disclosure documents are substantially harder. The results therefore do not support a...

Related works by Yonathan Arbel:

Search aliases:

Claim Annotations

No author-reviewed claim atoms are currently published for this paper.

Evidence-Linked Propositions

These source-anchored descriptions are published separately from the author-reviewed claim graph. Check each record’s review status.

the plain-language movement produced nearly 800 laws while resting on an asserted scientific readability crisis

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 1, that the plain-language movement produced nearly 800 laws while resting on an asserted scientific readability crisis. The discussion situates this proposition within Abstract and the article's empirical reassessment of the plain-language movement. This is significant because a largely technocratic reform project achieved enormous legal reach without comparable public scrutiny of its empirical foundations. It connects to plain-language movement, consumer law, readability crisis, technocratic regulation, empirical foundations, legal reform.

printed pp. 1 (PDF pp. 1) · Review: machine-drafted-source-checked

the article supplies the largest empirical test of plain-language claims by analyzing roughly two million contracts across industries and decades

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 1, that the article supplies the largest empirical test of plain-language claims by analyzing roughly two million contracts across industries and decades. The discussion situates this proposition within Abstract and the article's empirical reassessment of the plain-language movement. This is significant because scale and longitudinal breadth permit stronger benchmarking than studies built from a few hundred selected documents. It connects to big data, contract corpus, empirical legal studies, longitudinal analysis, machine learning, benchmarking.

printed pp. 1 (PDF pp. 1) · Review: machine-drafted-source-checked

consumer agreements have median readability scores close to those of daily news articles

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 1, that consumer agreements have median readability scores close to those of daily news articles. The discussion situates this proposition within Abstract and the article's empirical reassessment of the plain-language movement. This is significant because the central empirical premise of a general consumer-contract readability crisis is not supported by the study's benchmark. It connects to consumer contracts, daily news, comparative readability, empirical benchmark, plain language, consumer protection.

printed pp. 1 (PDF pp. 1) · Review: machine-drafted-source-checked

standard readability tools are unreliable, manipulable, and can differ by up to 4.6 grade levels on identical text

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 1, that standard readability tools are unreliable, manipulable, and can differ by up to 4.6 grade levels on identical text. The discussion situates this proposition within Abstract and the article's empirical reassessment of the plain-language movement. This is significant because laws and compliance systems that treat a score as objective can reward implementation choice rather than genuine accessibility. It connects to readability formulas, measurement error, manipulability, Flesch-Kincaid, regulatory metrics, validity.

printed pp. 1 (PDF pp. 1) · Review: machine-drafted-source-checked

the belief that most American adults cannot read beyond eighth grade is an unsupported myth

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 1, that the belief that most American adults cannot read beyond eighth grade is an unsupported myth. The discussion situates this proposition within Abstract and the article's empirical reassessment of the plain-language movement. This is significant because a false literacy baseline can misdirect consumer policy and obscure the needs of genuinely vulnerable readers. It connects to adult literacy, eighth-grade myth, consumer policy, vulnerable populations, evidence-based law, plain language.

printed pp. 1 (PDF pp. 1) · Review: machine-drafted-source-checked

consumer protection should shift from superficial readability scores toward market structure, substantive terms, and targeted assistance

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 1, that consumer protection should shift from superficial readability scores toward market structure, substantive terms, and targeted assistance. The discussion situates this proposition within Abstract and the article's empirical reassessment of the plain-language movement. This is significant because clear prose cannot cure exploitative bargains or structural failures, while personalized support can address actual reader needs. It connects to market structure, substantive fairness, targeted assistance, consumer vulnerability, smart readers, regulatory priorities.

printed pp. 1 (PDF pp. 1) · Review: machine-drafted-source-checked

the no-reading problem cuts against contract law's organizing ideal of assent

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 2, that the no-reading problem cuts against contract law's organizing ideal of assent. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because people who do not know terms may accept unfavorable deals and supply only paper-thin consent. It connects to no-reading problem, assent, contract theory, consumer welfare, autonomy, standard forms.

printed pp. 2 (PDF pp. 2) · Review: machine-drafted-source-checked

consumers decline to read for many reasons unrelated to linguistic difficulty

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 2, that consumers decline to read for many reasons unrelated to linguistic difficulty. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because nonnegotiability, trust, overload, timing, boredom, and the gap between written terms and firm behavior all weaken incentives to read. It connects to nonnegotiability, trust, information overload, firm behavior, reading incentives, consumer contracts.

printed pp. 2 (PDF pp. 2) · Review: machine-drafted-source-checked

legal scholarship often treats arcane language as the leading cause of nonreading

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 2, that legal scholarship often treats arcane language as the leading cause of nonreading. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because the plain-language agenda depends on a causal bridge from textual complexity to consumer disengagement and harm. It connects to legalese, causal inference, nonreading, consumer harm, plain language, contract design.

printed pp. 2 (PDF pp. 2) · Review: machine-drafted-source-checked

the claimed readability gap is used to justify protection, competition, and informed-rights rationales

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 2, that the claimed readability gap is used to justify protection, competition, and informed-rights rationales. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because if the gap or the causal effects are misstated, multiple strands of consumer policy lose their empirical footing. It connects to readability gap, consumer protection, competition, rights knowledge, policy justification, empirical support.

printed pp. 2 (PDF pp. 2) · Review: machine-drafted-source-checked

the plain-language movement is politically quiet but legally powerful

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 3, that the plain-language movement is politically quiet but legally powerful. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because its limited public visibility masks extensive influence on lending, warranties, insurance, state contracts, and unfair-practice regulation. It connects to political visibility, Truth in Lending, warranties, insurance regulation, state law, consumer protection.

printed pp. 3 (PDF pp. 3) · Review: machine-drafted-source-checked

nearly 800 federal and state plain-language laws amount to a major consumer-protection revolution

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 3, that nearly 800 federal and state plain-language laws amount to a major consumer-protection revolution. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because the scale of enacted policy makes evaluation of effectiveness and opportunity cost urgent. It connects to statutory diffusion, consumer revolution, policy evaluation, regulatory scale, plain language, opportunity cost.

printed pp. 3 (PDF pp. 3) · Review: machine-drafted-source-checked

the article tests four propositions: comparative readability, change over time, metric reliability, and adult literacy

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 3, that the article tests four propositions: comparative readability, change over time, metric reliability, and adult literacy. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because separating these questions prevents a single intuition about difficult legal prose from standing in for multiple empirical claims. It connects to research design, comparative readability, time trends, measurement reliability, adult literacy, hypothesis testing.

printed pp. 3 (PDF pp. 3) · Review: machine-drafted-source-checked

benchmarking contracts against texts adults voluntarily read is more informative than interpreting grade levels in isolation

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 3, that benchmarking contracts against texts adults voluntarily read is more informative than interpreting grade levels in isolation. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because a relative comparison avoids treating a contested formula output as a literal measure of human capacity. It connects to benchmarking, grade-level scores, daily reading, measurement interpretation, consumer behavior, comparative method.

printed pp. 3 (PDF pp. 3) · Review: machine-drafted-source-checked

the article contributes both a two-million-document corpus and machine-learning methods for contract research

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 3, that the article contributes both a two-million-document corpus and machine-learning methods for contract research. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because methodological infrastructure can support future empirical work beyond the present readability dispute. It connects to research infrastructure, contract dataset, legal NLP, machine learning, open data, empirical methods.

printed pp. 3 (PDF pp. 3) · Review: machine-drafted-source-checked

a universal and mathematically provable ideal of plain language should be rejected

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 4, that a universal and mathematically provable ideal of plain language should be rejected. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because reader diversity and weak measurement make one-size-fits-all optimization conceptually unsound. It connects to universal metrics, reader heterogeneity, plain language, measurement limits, contract drafting, consumer diversity.

printed pp. 4 (PDF pp. 4) · Review: machine-drafted-source-checked

plain-language regulation can divert attention from substantive terms and market conditions to drafting compliance

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 4, that plain-language regulation can divert attention from substantive terms and market conditions to drafting compliance. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because a focus on form can consume regulatory resources while leaving unfair prices or power structures untouched. It connects to regulatory diversion, substantive terms, market structure, compliance costs, consumer harm, legal form.

printed pp. 4 (PDF pp. 4) · Review: machine-drafted-source-checked

plain-language policy can entrench a moralized duty to read and strict textualism

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 4, that plain-language policy can entrench a moralized duty to read and strict textualism. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because making forms nominally readable may strengthen doctrines that blame consumers for failing to absorb standardized terms. It connects to duty to read, textualism, consumer blame, shrinkwrap, contract doctrine, plain language.

printed pp. 4 (PDF pp. 4) · Review: machine-drafted-source-checked

personalized smart readers offer a better direction than drafting for an imagined average consumer

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 4, that personalized smart readers offer a better direction than drafting for an imagined average consumer. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because technology can tailor language, presentation, and modality to actual linguistic and cognitive needs. It connects to smart readers, personalization, accessibility, consumer diversity, legal technology, contract comprehension.

printed pp. 4 (PDF pp. 4) · Review: machine-drafted-source-checked

the plain-language story illustrates how folk empiricism can drive well-intentioned legal reform

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 4, that the plain-language story illustrates how folk empiricism can drive well-intentioned legal reform. The discussion situates this proposition within Introduction and the article's empirical reassessment of the plain-language movement. This is significant because easy-to-repeat empirical myths can redirect law and scholarship away from interventions with stronger evidence. It connects to folk empiricism, legal myths, evidence-based policy, consumer reform, ALL-CAPS, scholarly rigor.

printed pp. 4 (PDF pp. 4) · Review: machine-drafted-source-checked

the plain-language movement combines political invisibility, elite endorsement, populist rhetoric, and global reach

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 5, that the plain-language movement combines political invisibility, elite endorsement, populist rhetoric, and global reach. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because its paradoxical coalition helps explain how a low-salience project achieved durable bipartisan influence. It connects to political coalition, populism, elite endorsement, global diffusion, plain language, regulatory history.

printed pp. 5 (PDF pp. 5) · Review: machine-drafted-source-checked

the movement emerged when substantive consumer regulation weakened and process-based reform looked more attainable

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 5, that the movement emerged when substantive consumer regulation weakened and process-based reform looked more attainable. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because historical context may have pushed advocates toward technocratic drafting rules instead of contested questions of value and power. It connects to regulatory history, FTC, process reform, substantive regulation, technocracy, consumer politics.

printed pp. 5 (PDF pp. 5) · Review: machine-drafted-source-checked

literacy anxiety and the Nation at Risk narrative helped shape the movement's view of Americans as deeply undereducated

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 5, that literacy anxiety and the Nation at Risk narrative helped shape the movement's view of Americans as deeply undereducated. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because a false or exaggerated diagnosis of schooling can become embedded in later disclosure policy. It connects to literacy anxiety, Nation at Risk, education myths, policy diffusion, adult readers, plain language.

printed pp. 5 (PDF pp. 5) · Review: machine-drafted-source-checked

the movement's manuals and style guides conceal jurisprudential and political commitments beneath craft advice

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 5, that the movement's manuals and style guides conceal jurisprudential and political commitments beneath craft advice. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because choices about legal form, access, and obligation cannot be reduced to preferences between short and long phrases. It connects to legal style, jurisprudence, political commitments, technocracy, legal realism, drafting manuals.

printed pp. 5 (PDF pp. 5) · Review: machine-drafted-source-checked

plain-language reform has attracted bipartisan political support and civil-rights rhetoric

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 6, that plain-language reform has attracted bipartisan political support and civil-rights rhetoric. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because broad rhetorical appeal helps explain statutory success without resolving whether the policies work. It connects to bipartisanship, Plain Writing Act, civil rights, political rhetoric, regulatory adoption, policy efficacy.

printed pp. 6 (PDF pp. 6) · Review: machine-drafted-source-checked

the claim that people do not read contracts at all is exaggerated

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 6, that the claim that people do not read contracts at all is exaggerated. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because consumers sometimes read, ask about policies, use reputational information, and rely on market signals outside the formal document. It connects to consumer reading, reputation markets, warranties, information channels, empirical behavior, no-reading problem.

printed pp. 6 (PDF pp. 6) · Review: machine-drafted-source-checked

even demonstrated nonreading of online terms does not establish that reading consistently improves consumer outcomes

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 6, that even demonstrated nonreading of online terms does not establish that reading consistently improves consumer outcomes. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the welfare value of disclosure depends on actionability and decision effects, not exposure to text alone. It connects to online terms, disclosure efficacy, consumer decisions, welfare, attention, contract reading.

printed pp. 6 (PDF pp. 6) · Review: machine-drafted-source-checked

legalese is only one of several plausible obstacles to reading

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 6, that legalese is only one of several plausible obstacles to reading. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because a nonnegotiable document may remain irrational to read even after excellent copyediting. It connects to legalese, nonnegotiability, rational inattention, reading costs, causal mechanisms, consumer contracts.

printed pp. 6 (PDF pp. 6) · Review: machine-drafted-source-checked

concerns about unreadable boilerplate have supported nonenforcement, enhanced disclosure, and substantive policing

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 7, that concerns about unreadable boilerplate have supported nonenforcement, enhanced disclosure, and substantive policing. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because an empirical claim about language supplies leverage for much broader doctrinal responses. It connects to boilerplate, nonenforcement, conspicuous disclosure, substantive regulation, consumer consent, doctrinal implications.

printed pp. 7 (PDF pp. 7) · Review: machine-drafted-source-checked

plain-language advocates generally characterize legalese as poor craft rather than strategic obfuscation

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 7, that plain-language advocates generally characterize legalese as poor craft rather than strategic obfuscation. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the diagnosis places responsibility on drafting technique and promises reform without confronting distributional conflict. It connects to bad drafting, strategic obfuscation, lawyer incentives, craft, distributional conflict, plain language.

printed pp. 7 (PDF pp. 7) · Review: machine-drafted-source-checked

translation from technical legal language to plain language can be lossy

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 7, that translation from technical legal language to plain language can be lossy. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because precision sometimes requires distinctions that a compressed or simplified version may erase. It connects to lossy compression, legal precision, technical terms, information theory, contract interpretation, simplification.

printed pp. 7 (PDF pp. 7) · Review: machine-drafted-source-checked

the case for plain language cannot assume that simplicity and precision are costless complements

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 7, that the case for plain language cannot assume that simplicity and precision are costless complements. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the relevant policy question is how much meaning, detail, and length a redesign changes. It connects to simplicity, precision, tradeoffs, document design, legal meaning, consumer disclosure.

printed pp. 7 (PDF pp. 7) · Review: machine-drafted-source-checked

explaining technical terms in simpler words often lengthens the document

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 8, that explaining technical terms in simpler words often lengthens the document. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because improvement along sentence complexity can worsen the separate burden of reading time and volume. It connects to length-simplicity tradeoff, document burden, reading time, technical explanation, insurance contracts, multi-dimensional readability.

printed pp. 8 (PDF pp. 8) · Review: machine-drafted-source-checked

Dutch insurance simplification produced documents that were easier by one metric but substantially longer

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 8, that Dutch insurance simplification produced documents that were easier by one metric but substantially longer. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because a policy can look successful on a chosen score while leaving total accessibility unchanged or worse. It connects to insurance policies, metric tradeoffs, document length, policy evaluation, readability, unintended effects.

printed pp. 8 (PDF pp. 8) · Review: machine-drafted-source-checked

the relationship between linguistic simplification and comprehension is not linear

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 8, that the relationship between linguistic simplification and comprehension is not linear. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because readers may prefer and understand text matched to their knowledge rather than the simplest possible prose. It connects to comprehension, reader-text match, nonlinearity, linguistic complexity, consumer preferences, plain language.

printed pp. 8 (PDF pp. 8) · Review: machine-drafted-source-checked

readability formulas operationalize a rich construct through a few formal textual counts

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 8, that readability formulas operationalize a rich construct through a few formal textual counts. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because word, sentence, character, syllable, and difficult-word features omit meaning, discourse, context, and reader characteristics. It connects to operationalization, formal metrics, linguistics, text features, construct validity, reader context.

printed pp. 8 (PDF pp. 8) · Review: machine-drafted-source-checked

widely used readability tests rely on different combinations of surface features

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 9, that widely used readability tests rely on different combinations of surface features. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because scores labeled readability may encode materially different constructs and assumptions. It connects to Flesch-Kincaid, SMOG, Gunning Fog, Coleman-Liau, surface features, measurement theory.

printed pp. 9 (PDF pp. 9) · Review: machine-drafted-source-checked

policymakers and commercial actors use readability scores despite advocates' own warnings that the measures are reductive

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 9, that policymakers and commercial actors use readability scores despite advocates' own warnings that the measures are reductive. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because a hedged research tool becomes dangerous when converted into a compliance threshold. It connects to regulatory thresholds, measurement caution, compliance, readability software, policy implementation, Goodhart's law.

printed pp. 9 (PDF pp. 9) · Review: machine-drafted-source-checked

readability measures require separate evaluation of internal and external validity

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 9, that readability measures require separate evaluation of internal and external validity. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because agreement among tests and correlation with real comprehension are distinct questions that policy must not conflate. It connects to internal validity, external validity, measurement theory, comprehension, reliability, policy evidence.

printed pp. 9 (PDF pp. 9) · Review: machine-drafted-source-checked

readability formulas that purport to measure one construct should correlate on a common text corpus

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 10, that readability formulas that purport to measure one construct should correlate on a common text corpus. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because inter-test agreement is a basic prediction of a unified readability concept. It connects to correlation, construct validity, Gutenberg corpus, measurement consistency, readability formulas, empirical test.

printed pp. 10 (PDF pp. 10) · Review: machine-drafted-source-checked

several leading readability tests correlate weakly or negatively with one another

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 10, that several leading readability tests correlate weakly or negatively with one another. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the label readability does not guarantee that different formulas measure the same property. It connects to correlation matrix, SMOG, Flesch-Kincaid, Coleman-Liau, construct fragmentation, metric validity.

printed pp. 10 (PDF pp. 10) · Review: machine-drafted-source-checked

near-perfect correlations among some tests suggest redundancy rather than independent confirmation

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 10, that near-perfect correlations among some tests suggest redundancy rather than independent confirmation. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because multiple formula outputs may give a false appearance of triangulation while repeating the same signal. It connects to redundancy, triangulation, Dale-Chall, Gunning Fog, measurement independence, readability.

printed pp. 10 (PDF pp. 10) · Review: machine-drafted-source-checked

different software libraries implementing Flesch-Kincaid differed by an average of 4.6 grade levels on the same contract paragraphs

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 10, that different software libraries implementing Flesch-Kincaid differed by an average of 4.6 grade levels on the same contract paragraphs. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because implementation choice alone can determine legal compliance or scholarly conclusions. It connects to software implementation, Flesch-Kincaid, 4.6-year gap, reproducibility, metric gaming, contract paragraphs.

printed pp. 10 (PDF pp. 10) · Review: machine-drafted-source-checked

sentence tokenization is ambiguous in contracts with headings, bullets, definitions, and conversion artifacts

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 11, that sentence tokenization is ambiguous in contracts with headings, bullets, definitions, and conversion artifacts. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because a mechanical parser can transform formatting into fantastical grade-level scores unrelated to reader difficulty. It connects to sentence tokenization, document formatting, conversion artifacts, headings, measurement error, legal documents.

printed pp. 11 (PDF pp. 11) · Review: machine-drafted-source-checked

word and syllable counts depend on contested linguistic and computational choices

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 11, that word and syllable counts depend on contested linguistic and computational choices. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because compound words, contractions, accents, URLs, and pronunciation expose hidden discretion inside supposedly objective formulas. It connects to word boundaries, syllable counting, pronunciation, URLs, computational linguistics, objectivity.

printed pp. 11 (PDF pp. 11) · Review: machine-drafted-source-checked

earlier readability research often overlooks implementation sensitivity

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 11, that earlier readability research often overlooks implementation sensitivity. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because replication and regulatory use require disclosure of software, preprocessing, and tokenization choices. It connects to method transparency, replication, preprocessing, software libraries, readability research, measurement sensitivity.

printed pp. 11 (PDF pp. 11) · Review: machine-drafted-source-checked

readability formulas show mixed and sometimes inverse relationships with actual comprehension

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 11, that readability formulas show mixed and sometimes inverse relationships with actual comprehension. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because formal ease cannot be assumed to improve learning, preference, or use without outcome validation. It connects to external validity, comprehension, reader preference, scientific abstracts, outcome validation, readability.

printed pp. 11 (PDF pp. 11) · Review: machine-drafted-source-checked

classic formulas explain only a small share of measured comprehension and reading-speed variance in recent studies

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 12, that classic formulas explain only a small share of measured comprehension and reading-speed variance in recent studies. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because modest correlations are insufficient to support precise grade-level mandates. It connects to explained variance, reading speed, comprehension, Wikipedia, grade mandates, predictive validity.

printed pp. 12 (PDF pp. 12) · Review: machine-drafted-source-checked

contract-simplification experiments produce mixed or null effects on understanding, trust, and behavior

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 12, that contract-simplification experiments produce mixed or null effects on understanding, trust, and behavior. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because copyediting cannot be presumed to change consumer choices or dispute responses. It connects to contract experiments, simplification, trust, consumer behavior, comprehension, causal evidence.

printed pp. 12 (PDF pp. 12) · Review: machine-drafted-source-checked

some experiments found the nominally least readable forms produced the most correct answers

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 12, that some experiments found the nominally least readable forms produced the most correct answers. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because formula scores may diverge from task-specific comprehension in legally meaningful settings. It connects to correct answers, contract forms, formula validity, task-specific comprehension, counterintuitive evidence, consumer research.

printed pp. 12 (PDF pp. 12) · Review: machine-drafted-source-checked

making contracts shorter may improve comprehension more reliably than changing formula scores

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 12, that making contracts shorter may improve comprehension more reliably than changing formula scores. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because document volume is a distinct access problem that sentence-level metrics often miss. It connects to contract length, shortening, comprehension, document design, reading burden, multi-dimensional access.

printed pp. 12 (PDF pp. 12) · Review: machine-drafted-source-checked

a large experiment found easier contracts felt more understandable but did not change advice-seeking, complaints, or legal action

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 12, that a large experiment found easier contracts felt more understandable but did not change advice-seeking, complaints, or legal action. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because subjective clarity and consequential behavior should be measured separately. It connects to subjective understanding, behavioral outcomes, legal action, advice seeking, insurance contracts, experimental evidence.

printed pp. 12 (PDF pp. 12) · Review: machine-drafted-source-checked

claims that half of American adults cannot read past eighth grade are pervasive across scholarship and government materials

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 13, that claims that half of American adults cannot read past eighth grade are pervasive across scholarship and government materials. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because repetition across institutions can create authority without a sound originating source. It connects to citation cascade, adult literacy, government reports, legal scholarship, eighth-grade claim, epistemic reliability.

printed pp. 13 (PDF pp. 13) · Review: machine-drafted-source-checked

if most adults truly read like young teenagers, the implication would reach democracy and innovation far beyond contracts

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 13, that if most adults truly read like young teenagers, the implication would reach democracy and innovation far beyond contracts. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the extraordinary breadth of the claim should trigger correspondingly rigorous validation. It connects to extraordinary claims, democracy, innovation, literacy, policy stakes, evidence standards.

printed pp. 13 (PDF pp. 13) · Review: machine-drafted-source-checked

Census educational-attainment data contradicts the premise of population-wide arrest at middle-school reading levels

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 13, that Census educational-attainment data contradicts the premise of population-wide arrest at middle-school reading levels. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because formal attainment cannot prove comprehension but supplies a powerful plausibility check against the myth. It connects to Census, educational attainment, plausibility check, adult readers, literacy myth, population data.

printed pp. 13 (PDF pp. 13) · Review: machine-drafted-source-checked

adult literacy should not be inferred by mechanically equating formula grade levels with completed schooling

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 13, that adult literacy should not be inferred by mechanically equating formula grade levels with completed schooling. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the two scales describe different constructs and their numerical resemblance invites category error. It connects to grade-level formulas, schooling, construct mismatch, adult literacy, measurement interpretation, category error.

printed pp. 13 (PDF pp. 13) · Review: machine-drafted-source-checked

PIAAC finds most American adults have mid or high English proficiency

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 14, that PIAAC finds most American adults have mid or high English proficiency. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because direct skill assessment does not support describing a majority as functionally or partially illiterate. It connects to PIAAC, adult proficiency, direct assessment, functional literacy, international comparison, education evidence.

printed pp. 14 (PDF pp. 14) · Review: machine-drafted-source-checked

NAAL expressly warns that grade-level translations of literacy are arbitrary

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 14, that NAAL expressly warns that grade-level translations of literacy are arbitrary. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because sources invoked for the eighth-grade claim undermine rather than substantiate that translation. It connects to NAAL, arbitrary standards, source verification, literacy scales, citation accuracy, grade equivalents.

printed pp. 14 (PDF pp. 14) · Review: machine-drafted-source-checked

people classified below a demanding literacy threshold can still perform complex document tasks

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 14, that people classified below a demanding literacy threshold can still perform complex document tasks. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because categorical labels like partially illiterate obscure functional capacities and threshold choices. It connects to literacy thresholds, functional capacity, document comparison, classification, public rhetoric, measurement.

printed pp. 14 (PDF pp. 14) · Review: machine-drafted-source-checked

a Gallup report created an alarming partial-illiteracy figure by defining everyone below Level 3 as partially illiterate

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 14, that a Gallup report created an alarming partial-illiteracy figure by defining everyone below Level 3 as partially illiterate. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the headline depends on a stipulative label not used by the underlying international assessment. It connects to Gallup, PIAAC levels, definitional inflation, partial illiteracy, source interpretation, public statistics.

printed pp. 14 (PDF pp. 14) · Review: machine-drafted-source-checked

self-reported and educational evidence should be read alongside direct assessments rather than collapsed into a single grade number

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 14, that self-reported and educational evidence should be read alongside direct assessments rather than collapsed into a single grade number. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because triangulation reveals reader capacity more faithfully than one unsupported conversion. It connects to triangulation, self-report, education data, direct assessment, literacy, measurement pluralism.

printed pp. 14 (PDF pp. 14) · Review: machine-drafted-source-checked

health-literacy studies arise in contexts of stress, disability, and specialized medical material

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 15, that health-literacy studies arise in contexts of stress, disability, and specialized medical material. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because findings from small clinical samples should not be generalized uncritically to all adults and contracts. It connects to health literacy, external validity, clinical context, sample size, medical disclosures, generalization.

printed pp. 15 (PDF pp. 15) · Review: machine-drafted-source-checked

a central grade-equivalence assertion traces to an unsourced rough approximation in an old textbook

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 15, that a central grade-equivalence assertion traces to an unsourced rough approximation in an old textbook. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because citation chains can transform a tentative teaching heuristic into an apparent empirical fact. It connects to citation chain, rough approximation, textbook, grade equivalence, legal myths, source audit.

printed pp. 15 (PDF pp. 15) · Review: machine-drafted-source-checked

book, news, and social-media consumption provide a sanity check against majority illiteracy

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 15, that book, news, and social-media consumption provide a sanity check against majority illiteracy. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because observed voluntary interaction with nontrivial text is inconsistent with the movement's most sweeping premise. It connects to reading habits, book consumption, news, social media, sanity check, adult literacy.

printed pp. 15 (PDF pp. 15) · Review: machine-drafted-source-checked

most eighth and twelfth graders read at or above their own basic grade level

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 15, that most eighth and twelfth graders read at or above their own basic grade level. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because the claim that half of adults never progress beyond eighth grade is implausible even against school-age benchmarks. It connects to student achievement, eighth graders, twelfth graders, basic reading, adult literacy, plausibility.

printed pp. 15 (PDF pp. 15) · Review: machine-drafted-source-checked

policy should focus on people with real literacy and accessibility vulnerabilities rather than an invented average deficit

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 16, that policy should focus on people with real literacy and accessibility vulnerabilities rather than an invented average deficit. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because rejecting a population-wide myth strengthens rather than weakens the case for targeted support. It connects to targeted policy, literacy vulnerability, accessibility, consumer protection, average consumer, distributional focus.

printed pp. 16 (PDF pp. 16) · Review: machine-drafted-source-checked

readers who struggle with contracts are heterogeneous in language, cognition, vision, education, age, and context

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 16, that readers who struggle with contracts are heterogeneous in language, cognition, vision, education, age, and context. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because a single grade-level score cannot specify the intervention that each person needs. It connects to reader heterogeneity, neurodivergence, language access, visual accessibility, context, personalization.

printed pp. 16 (PDF pp. 16) · Review: machine-drafted-source-checked

false assumptions about the average reader produce false expectations and misdirected policy

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 16, that false assumptions about the average reader produce false expectations and misdirected policy. The discussion situates this proposition within I. The Four Tenets of Plain Language and the article's empirical reassessment of the plain-language movement. This is significant because scientific rigor matters because measurement errors become embedded in broad legal mandates. It connects to average reader, policy design, scientific rigor, measurement error, legal mandates, plain language.

printed pp. 16 (PDF pp. 16) · Review: machine-drafted-source-checked

the study tests the movement's central readability claim despite serious doubts about traditional metrics

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 17, that the study tests the movement's central readability claim despite serious doubts about traditional metrics. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because robust preprocessing, aggregation, and benchmarking allow a conservative test using the movement's own tools. It connects to adversarial testing, robust metrics, benchmarking, plain-language claims, empirical design, readability.

printed pp. 17 (PDF pp. 17) · Review: machine-drafted-source-checked

SEC exhibit-10 filings provide a corpus of roughly 1.2 million material commercial contracts

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 17, that SEC exhibit-10 filings provide a corpus of roughly 1.2 million material commercial contracts. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because mandatory disclosure creates longitudinal legal data at a scale unavailable through conventional manual collection. It connects to SEC EDGAR, Exhibit 10, material contracts, mandatory disclosure, commercial agreements, big data.

printed pp. 17 (PDF pp. 17) · Review: machine-drafted-source-checked

format normalization is essential because parsing artifacts can distort readability scores

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 17, that format normalization is essential because parsing artifacts can distort readability scores. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because measurement validity depends on cleaning HTML, PDF, and plain-text sources into comparable documents. It connects to data cleaning, PDF conversion, HTML parsing, format artifacts, readability measurement, reproducibility.

printed pp. 17 (PDF pp. 17) · Review: machine-drafted-source-checked

manual classification is infeasible for a million-contract corpus

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 17, that manual classification is infeasible for a million-contract corpus. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because legal NLP is necessary to separate heterogeneous agreement types before meaningful comparison. It connects to document classification, legal NLP, scale, contract taxonomy, machine learning, empirical law.

printed pp. 17 (PDF pp. 17) · Review: machine-drafted-source-checked

hundreds of hand-coded title rules classified only about one quarter of the contracts

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 18, that hundreds of hand-coded title rules classified only about one quarter of the contracts. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because legal drafting variation and idiosyncratic labels limit brittle rule-based systems. It connects to rule-based classification, contract titles, drafting variation, coverage, legal language, automation.

printed pp. 18 (PDF pp. 18) · Review: machine-drafted-source-checked

contract titles can supply weak labels created by the drafting lawyers themselves

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 18, that contract titles can supply weak labels created by the drafting lawyers themselves. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because built-in metadata avoids prohibitively expensive manual annotation while retaining legally informed categories. It connects to weak supervision, built-in labels, contract titles, annotation cost, legal expertise, training data.

printed pp. 18 (PDF pp. 18) · Review: machine-drafted-source-checked

titles were grouped into ten categories and removed before model training

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 18, that titles were grouped into ten categories and removed before model training. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because scrubbing the label forces the classifier to learn substantive content rather than a superficial title cue. It connects to label leakage, title scrubbing, contract categories, substantive content, machine learning, model design.

printed pp. 18 (PDF pp. 18) · Review: machine-drafted-source-checked

the weak-label process produced 275,480 training examples across leading contract categories

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 18, that the weak-label process produced 275,480 training examples across leading contract categories. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because large labeled data makes high-coverage classification possible without pretending the categories exhaust every agreement type. It connects to training corpus, contract taxonomy, weak labels, sample size, classification coverage, legal datasets.

printed pp. 18 (PDF pp. 18) · Review: machine-drafted-source-checked

traditional TF-IDF vectors outperformed transformer embeddings for contract-type classification

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 19, that traditional TF-IDF vectors outperformed transformer embeddings for contract-type classification. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because newer and more computationally intensive representations are not automatically superior for domain tasks driven by distinctive vocabulary. It connects to TF-IDF, transformers, BERT, legal BERT, model comparison, domain vocabulary.

printed pp. 19 (PDF pp. 19) · Review: machine-drafted-source-checked

TF-IDF works well because rare specialized words distinguish contract categories

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 19, that TF-IDF works well because rare specialized words distinguish contract categories. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because terms such as mortgage, salary, collateral, and termination carry more class information than ubiquitous words. It connects to feature importance, specialized vocabulary, document vectors, contract types, information retrieval, TF-IDF.

printed pp. 19 (PDF pp. 19) · Review: machine-drafted-source-checked

LightGBM supplied the best combination of accuracy and inference speed among tested classifiers

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 19, that LightGBM supplied the best combination of accuracy and inference speed among tested classifiers. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because large-corpus analysis requires attention to computational cost as well as headline accuracy. It connects to LightGBM, gradient boosting, inference time, classification accuracy, computational efficiency, model selection.

printed pp. 19 (PDF pp. 19) · Review: machine-drafted-source-checked

complex models and embeddings imposed substantial computing costs without winning the task

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 19, that complex models and embeddings imposed substantial computing costs without winning the task. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because empirical legal research benefits from testing simple baselines before adopting resource-intensive architectures. It connects to simple baselines, GPU computing, model efficiency, legal NLP, research design, cost-benefit analysis.

printed pp. 19 (PDF pp. 19) · Review: machine-drafted-source-checked

the classifier reached 96.67 percent balanced accuracy within the defined training categories

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 20, that the classifier reached 96.67 percent balanced accuracy within the defined training categories. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because high held-out performance supports large-scale descriptive use when its domain boundary is respected. It connects to balanced accuracy, held-out test, contract classification, model performance, domain boundary, legal NLP.

printed pp. 20 (PDF pp. 20) · Review: machine-drafted-source-checked

category-level F1 scores ranged from roughly 93 to 100 percent for some agreement types

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 20, that category-level F1 scores ranged from roughly 93 to 100 percent for some agreement types. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because aggregate accuracy should be supplemented with class-specific performance to detect uneven errors. It connects to F1 score, class imbalance, category performance, evaluation metrics, contract taxonomy, model audit.

printed pp. 20 (PDF pp. 20) · Review: machine-drafted-source-checked

misclassification between credit and purchase agreements is substantively intelligible

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 20, that misclassification between credit and purchase agreements is substantively intelligible. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because error analysis can reveal overlapping legal content rather than random model failure. It connects to confusion matrix, credit agreements, purchase agreements, error analysis, legal overlap, model interpretation.

printed pp. 20 (PDF pp. 20) · Review: machine-drafted-source-checked

a blind audit outside the structured test set produced a lower 78 percent balanced accuracy

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 20, that a blind audit outside the structured test set produced a lower 78 percent balanced accuracy. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because real-world generalization is materially weaker than performance on familiar categories and should bound downstream claims. It connects to blind audit, generalization, distribution shift, 78 percent, model validation, epistemic caution.

printed pp. 20 (PDF pp. 20) · Review: machine-drafted-source-checked

the CARD Act created a near-marketwide database of consumer credit-card agreements

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 21, that the CARD Act created a near-marketwide database of consumer credit-card agreements. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because regulatory filing duties can generate unusually representative data about standardized consumer contracts. It connects to CARD Act, CFPB, credit-card agreements, regulatory data, consumer contracts, market coverage.

printed pp. 21 (PDF pp. 21) · Review: machine-drafted-source-checked

credit-card contracts matter at population scale because most Americans use multiple cards

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 21, that credit-card contracts matter at population scale because most Americans use multiple cards. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because a standardized form in the dataset can govern millions of individual transactions. It connects to credit cards, population exposure, standard forms, consumer finance, contract prevalence, representativeness.

printed pp. 21 (PDF pp. 21) · Review: machine-drafted-source-checked

the CFPB files required extensive repair because of missing mappings, duplicates, scans, and corrupt documents

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 21, that the CFPB files required extensive repair because of missing mappings, duplicates, scans, and corrupt documents. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because public availability does not eliminate data-engineering and quality-assurance burdens. It connects to data quality, CFPB, PDF repair, duplicates, scanned documents, quality assurance.

printed pp. 21 (PDF pp. 21) · Review: machine-drafted-source-checked

quality control yielded 93,078 readable credit agreements from 95,071 deduplicated files

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 21, that quality control yielded 93,078 readable credit agreements from 95,071 deduplicated files. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because transparent attrition documents the difference between nominal archives and analyzable data. It connects to sample attrition, credit agreements, deduplication, processable files, data pipeline, transparency.

printed pp. 21 (PDF pp. 21) · Review: machine-drafted-source-checked

the privacy-policy corpus uses historical web snapshots for more than half a million sites

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 22, that the privacy-policy corpus uses historical web snapshots for more than half a million sites. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because archived longitudinal data permits comparison of drafting trends outside conventional filed contracts. It connects to privacy policies, Wayback Machine, Alexa rankings, longitudinal web data, consumer documents, data curation.

printed pp. 22 (PDF pp. 22) · Review: machine-drafted-source-checked

missing-year privacy policies were filled forward under an assumption of no interim change

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 22, that missing-year privacy policies were filled forward under an assumption of no interim change. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because longitudinal scale requires an explicit imputation rule that may smooth real variation. It connects to imputation, missing data, privacy policies, longitudinal analysis, assumptions, measurement limits.

printed pp. 22 (PDF pp. 22) · Review: machine-drafted-source-checked

franchise-disclosure registries supplied 12,602 documents across multiple states

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 22, that franchise-disclosure registries supplied 12,602 documents across multiple states. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because a hybrid consumer-commercial setting can be studied through decentralized public filing systems. It connects to franchise disclosure, state registries, FDD, hybrid contracts, data scraping, consumer business boundary.

printed pp. 22 (PDF pp. 22) · Review: machine-drafted-source-checked

franchise disclosures are legally mandated because prospective franchisees face information and power concerns

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 22, that franchise disclosures are legally mandated because prospective franchisees face information and power concerns. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because their readability has special policy relevance despite the nominal business status of the reader. It connects to franchisees, mandatory disclosure, information asymmetry, small business, consumer protection, readability.

printed pp. 22 (PDF pp. 22) · Review: machine-drafted-source-checked

the final corpus contains 1,935,680 contracts, including 1.2 million commercial and 735,680 consumer or hybrid agreements

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 23, that the final corpus contains 1,935,680 contracts, including 1.2 million commercial and 735,680 consumer or hybrid agreements. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because the study's claims rest on an unusually broad empirical base with multiple market settings. It connects to corpus size, commercial contracts, consumer agreements, hybrid agreements, empirical scale, contract research.

printed pp. 23 (PDF pp. 23) · Review: machine-drafted-source-checked

counts of standardized forms understate the number of transactions and people governed

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 23, that counts of standardized forms understate the number of transactions and people governed. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because one credit-card form can represent tens of thousands of legally distinct agreements. It connects to standardized forms, transaction multiplicity, unit of analysis, credit agreements, population coverage, representativeness.

printed pp. 23 (PDF pp. 23) · Review: machine-drafted-source-checked

the consumer sample overrepresents contracts already subject to disclosure and clarity regulation

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 23, that the consumer sample overrepresents contracts already subject to disclosure and clarity regulation. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because regulatory selection may partly explain why observed credit agreements resemble ordinary news in difficulty. It connects to sample selection, regulated contracts, TILA, clarity mandates, credit cards, external validity.

printed pp. 23 (PDF pp. 23) · Review: machine-drafted-source-checked

social-media terms, retail purchases, and consumer-to-consumer agreements are not adequately represented

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 23, that social-media terms, retail purchases, and consumer-to-consumer agreements are not adequately represented. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because the results should not be generalized to every consumer contract category. It connects to missing contract types, social media, retail contracts, consumer-to-consumer, sample scope, generalization.

printed pp. 23 (PDF pp. 23) · Review: machine-drafted-source-checked

the study's limitations remain modest relative to the tiny selected samples used in prior plain-language research

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 23, that the study's limitations remain modest relative to the tiny selected samples used in prior plain-language research. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because empirical criticism should compare realistic evidence bases rather than demand perfection only from the revisionist study. It connects to comparative evidence, sample size, prior literature, research limitations, burden of proof, empirical rigor.

printed pp. 23 (PDF pp. 23) · Review: machine-drafted-source-checked

rigorous preprocessing removes nontextual artifacts and uses two sentence tokenizers

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 24, that rigorous preprocessing removes nontextual artifacts and uses two sentence tokenizers. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because readability estimates are highly sensitive to boundaries, so redundant parsing reduces catastrophic scoring errors. It connects to preprocessing, sentence tokenizers, spaCy, TextBlob, robustness, measurement error.

printed pp. 24 (PDF pp. 24) · Review: machine-drafted-source-checked

averaging multiple implementations of each formula constrains selective software choice

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 24, that averaging multiple implementations of each formula constrains selective software choice. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because within-test aggregation reduces researcher degrees of freedom and opportunistic compliance gaming. It connects to ensemble measurement, software implementations, researcher discretion, metric gaming, robustness, readability tests.

printed pp. 24 (PDF pp. 24) · Review: machine-drafted-source-checked

the Composite Readability Measure averages across tests as well as across implementations

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 24, that the Composite Readability Measure averages across tests as well as across implementations. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because a summary measure can capture multiple formal dimensions while exposing that no single formula is privileged. It connects to Composite Readability Measure, averaging, measurement ensemble, readability dimensions, robustness, metric design.

printed pp. 24 (PDF pp. 24) · Review: machine-drafted-source-checked

extreme and negative grade estimates were bounded to reduce the effect of pathological outputs

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 24, that extreme and negative grade estimates were bounded to reduce the effect of pathological outputs. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because robust analysis must prevent parser failures from dominating large-sample comparisons. It connects to outlier handling, score capping, negative grades, robust statistics, parser error, data cleaning.

printed pp. 24 (PDF pp. 24) · Review: machine-drafted-source-checked

the benchmark corpus contains roughly 1.8 million news articles, reviews, debates, books, and reference texts

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 25, that the benchmark corpus contains roughly 1.8 million news articles, reviews, debates, books, and reference texts. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because comparison across familiar genres anchors formula outputs in observed reading behavior. It connects to benchmark corpus, news, Amazon reviews, presidential debates, books, Wikipedia.

printed pp. 25 (PDF pp. 25) · Review: machine-drafted-source-checked

benchmark texts are consumed voluntarily by large and diverse audiences

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 25, that benchmark texts are consumed voluntarily by large and diverse audiences. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because revealed reading choices supply a stronger access baseline than unsupported claims about abstract grade capacity. It connects to voluntary reading, revealed behavior, audience diversity, daily news, adult literacy, comparative benchmark.

printed pp. 25 (PDF pp. 25) · Review: machine-drafted-source-checked

different benchmark genres attract different audiences and therefore provide a range rather than one universal standard

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 25, that different benchmark genres attract different audiences and therefore provide a range rather than one universal standard. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because heterogeneous references make the analysis comparative without recreating the mythical average reader. It connects to audience heterogeneity, genre effects, benchmark range, average reader, external validity, comparative method.

printed pp. 25 (PDF pp. 25) · Review: machine-drafted-source-checked

benchmarking does not erase the needs of readers with atypical linguistic or cognitive requirements

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 25, that benchmarking does not erase the needs of readers with atypical linguistic or cognitive requirements. The discussion situates this proposition within II. Data and Machine-Learning Classification and the article's empirical reassessment of the plain-language movement. This is significant because population comparisons and accessibility policy answer different questions and should be pursued together. It connects to atypical needs, accessibility, population benchmark, targeted support, consumer diversity, policy interpretation.

printed pp. 25 (PDF pp. 25) · Review: machine-drafted-source-checked

the empirical analysis separates cross-sectional comparisons from change over time

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 26, that the empirical analysis separates cross-sectional comparisons from change over time. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because current relative difficulty and historical trajectory are distinct claims about the success of plain-language policy. It connects to cross-sectional analysis, time trends, research design, contract readability, policy evaluation, longitudinal evidence.

printed pp. 26 (PDF pp. 26) · Review: machine-drafted-source-checked

the Composite Readability Measure is used as the principal cross-test summary rather than a literal schooling diagnosis

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 26, that the Composite Readability Measure is used as the principal cross-test summary rather than a literal schooling diagnosis. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because aggregated scores support ranking and comparison while preserving skepticism about grade-level interpretation. It connects to CRM, comparative ranking, grade levels, measurement caution, contract types, empirical findings.

printed pp. 26 (PDF pp. 26) · Review: machine-drafted-source-checked

Amazon reviews and presidential debates are among the easiest benchmark texts while commercial contracts and franchise disclosures are the hardest

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 27, that Amazon reviews and presidential debates are among the easiest benchmark texts while commercial contracts and franchise disclosures are the hardest. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because the data differentiates contract categories instead of treating legal documents as uniformly unreadable. It connects to genre hierarchy, Amazon reviews, presidential debates, commercial contracts, franchise disclosures, heterogeneity.

printed pp. 27 (PDF pp. 27) · Review: machine-drafted-source-checked

credit-card agreements sit near CNN and Fox News in median composite readability

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 27, that credit-card agreements sit near CNN and Fox News in median composite readability. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because a heavily regulated mass consumer contract is not meaningfully separated from ordinary news by the tested language metrics. It connects to credit-card agreements, CNN, Fox News, median readability, consumer finance, benchmarking.

printed pp. 27 (PDF pp. 27) · Review: machine-drafted-source-checked

privacy policies are somewhat harder than credit agreements but close to financial news and The New Republic

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 27, that privacy policies are somewhat harder than credit agreements but close to financial news and The New Republic. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because difficulty varies within consumer documents and may reflect market discipline as well as transaction complexity. It connects to privacy policies, financial news, New Republic, market discipline, consumer documents, comparative readability.

printed pp. 27 (PDF pp. 27) · Review: machine-drafted-source-checked

commercial contracts have a median composite grade estimate near 16.9 while franchise disclosures are near 15.74

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 27, that commercial contracts have a median composite grade estimate near 16.9 while franchise disclosures are near 15.74. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because sophisticated and hybrid transactions—not consumer credit—occupy the difficult end of the distribution. It connects to commercial contracts, franchise disclosures, CRM, grade estimates, sophistication, distribution.

printed pp. 27 (PDF pp. 27) · Review: machine-drafted-source-checked

consumer-contract distributions substantially overlap texts adults habitually and voluntarily read

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 28, that consumer-contract distributions substantially overlap texts adults habitually and voluntarily read. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because the central mass rather than isolated difficult tails should inform claims of a population-wide crisis. It connects to distribution overlap, central tendency, difficult tails, daily reading, consumer contracts, risk framing.

printed pp. 28 (PDF pp. 28) · Review: machine-drafted-source-checked

transactional complexity does not determine linguistic simplicity

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 28, that transactional complexity does not determine linguistic simplicity. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because credit agreements can be more complex transactions yet score easier than privacy policies. It connects to transaction complexity, linguistic complexity, credit agreements, privacy policies, document design, causal inference.

printed pp. 28 (PDF pp. 28) · Review: machine-drafted-source-checked

low market attention may help explain why privacy policies are harder than credit-card agreements

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 28, that low market attention may help explain why privacy policies are harder than credit-card agreements. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because regulation and market discipline can shape drafting independently of the underlying legal subject. It connects to market discipline, consumer inattention, privacy policies, credit regulation, drafting incentives, readability.

printed pp. 28 (PDF pp. 28) · Review: machine-drafted-source-checked

franchise disclosures raise special concern because formally commercial readers may resemble consumers in power and sophistication

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 28, that franchise disclosures raise special concern because formally commercial readers may resemble consumers in power and sophistication. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because the hardest documents appear in a market created around worries about small-business exploitation. It connects to franchisees, hybrid status, disclosure, power imbalance, small business, consumer protection.

printed pp. 28 (PDF pp. 28) · Review: machine-drafted-source-checked

a robust Flesch-Kincaid analysis reproduces the broad ordering found in the composite measure

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 29, that a robust Flesch-Kincaid analysis reproduces the broad ordering found in the composite measure. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because convergence across specifications strengthens the relative comparison even though absolute grade labels remain contestable. It connects to robustness check, Flesch-Kincaid, CRM, rank ordering, specification, empirical confidence.

printed pp. 29 (PDF pp. 29) · Review: machine-drafted-source-checked

document length must be analyzed separately from sentence-level readability

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 29, that document length must be analyzed separately from sentence-level readability. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because a contract can use accessible prose yet remain practically unreadable because it contains hundreds of pages. It connects to document length, reading burden, sentence complexity, multi-dimensional readability, contracts, time cost.

printed pp. 29 (PDF pp. 29) · Review: machine-drafted-source-checked

commercial contracts and franchise disclosures impose exceptionally large reading-time investments

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 30, that commercial contracts and franchise disclosures impose exceptionally large reading-time investments. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because high semantic accessibility cannot compensate for overwhelming volume. It connects to reading time, commercial contracts, franchise disclosures, document volume, consumer burden, accessibility.

printed pp. 30 (PDF pp. 30) · Review: machine-drafted-source-checked

consumer contracts are much longer than the news and review texts with similar grade scores

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 30, that consumer contracts are much longer than the news and review texts with similar grade scores. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because the finding against a linguistic crisis does not eliminate a separate length crisis. It connects to consumer contracts, news articles, length crisis, grade scores, attention costs, comparative readability.

printed pp. 30 (PDF pp. 30) · Review: machine-drafted-source-checked

contracts that score as readable remain long, contradicting the idea that plain language necessarily produces concision

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 31, that contracts that score as readable remain long, contradicting the idea that plain language necessarily produces concision. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because improvement on one measurement dimension may be offset by deterioration on another. It connects to concision, readability scores, tradeoffs, contract length, policy metrics, document design.

printed pp. 31 (PDF pp. 31) · Review: machine-drafted-source-checked

standard readability formulas ignore formatting even though formatting affects access

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 31, that standard readability formulas ignore formatting even though formatting affects access. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because surface linguistic scores omit typography, layout, headings, and visual structure. It connects to formatting, typography, layout, readability formulas, accessibility, document design.

printed pp. 31 (PDF pp. 31) · Review: machine-drafted-source-checked

ALL-CAPS is common in commercial and franchise documents and unusually prevalent in consumer credit agreements

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 31, that ALL-CAPS is common in commercial and franchise documents and unusually prevalent in consumer credit agreements. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because legal mandates and professional culture can degrade visual accessibility independently of prose difficulty. It connects to ALL-CAPS, consumer credit, commercial contracts, legal culture, format regulation, visual accessibility.

printed pp. 31 (PDF pp. 31) · Review: machine-drafted-source-checked

ALL-CAPS mandates may reflect a mistaken regulatory belief that capitalization helps consumers

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 31, that ALL-CAPS mandates may reflect a mistaken regulatory belief that capitalization helps consumers. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because legally compelled formatting can perpetuate a myth even when empirical research shows no readability gain. It connects to capitalization mandates, regulatory design, empirical myth, consumer friendly, formatting, ALL-CAPS.

printed pp. 31 (PDF pp. 31) · Review: machine-drafted-source-checked

credit contracts remain easier than privacy policies and commercial agreements across time

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 32, that credit contracts remain easier than privacy policies and commercial agreements across time. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because the cross-sectional hierarchy is not merely an artifact of one year. It connects to longitudinal robustness, credit contracts, privacy policies, commercial contracts, time trends, comparative difficulty.

printed pp. 32 (PDF pp. 32) · Review: machine-drafted-source-checked

commercial contracts became materially more complex from 2001 to 2022

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 32, that commercial contracts became materially more complex from 2001 to 2022. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because high-end drafting grew more formal despite technologies that made editing and reuse cheaper. It connects to commercial complexity, longitudinal trend, legal technology, formality, contract drafting, 2001-2022.

printed pp. 32 (PDF pp. 32) · Review: machine-drafted-source-checked

commercial-contract change provides evidence that the measurement system can detect real temporal movement

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 32, that commercial-contract change provides evidence that the measurement system can detect real temporal movement. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because stability in consumer scores cannot be dismissed simply as an insensitive metric. It connects to measurement sensitivity, temporal change, consumer stability, commercial contracts, robustness, trend analysis.

printed pp. 32 (PDF pp. 32) · Review: machine-drafted-source-checked

privacy policies became modestly easier after an initially difficult period

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 33, that privacy policies became modestly easier after an initially difficult period. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because a neglected improvement trend complicates narratives of steadily worsening consumer language. It connects to privacy policies, improvement, time trend, consumer language, regulation, empirical surprise.

printed pp. 33 (PDF pp. 33) · Review: machine-drafted-source-checked

consumer credit readability is comparatively easy but volatile around the 2015 data interruption

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 33, that consumer credit readability is comparatively easy but volatile around the 2015 data interruption. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because missing-year selection and administrative changes may explain apparent shifts. It connects to credit-card agreements, missing data, 2015 suspension, selection effects, time series, measurement caution.

printed pp. 33 (PDF pp. 33) · Review: machine-drafted-source-checked

consumer credit agreements show renewed complexity after 2020

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 33, that consumer credit agreements show renewed complexity after 2020. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because plain-language commitments do not guarantee a monotonic trajectory toward easier documents. It connects to post-2020, credit contracts, complexity trend, plain-language policy, longitudinal analysis, consumer finance.

printed pp. 33 (PDF pp. 33) · Review: machine-drafted-source-checked

noncompete and post-employment agreements grew roughly fourteen percent more difficult on the composite measure

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 35, that noncompete and post-employment agreements grew roughly fourteen percent more difficult on the composite measure. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because workers face increasing linguistic complexity in documents that govern mobility and livelihood. It connects to noncompetes, post-employment restrictions, workers, 14 percent, readability trend, labor contracts.

printed pp. 35 (PDF pp. 35) · Review: machine-drafted-source-checked

almost every commercial-contract category became more difficult over the study period

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 35, that almost every commercial-contract category became more difficult over the study period. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because rising complexity is broad rather than confined to one unusual transaction type. It connects to commercial categories, broad trend, contract complexity, longitudinal evidence, formalization, legal drafting.

printed pp. 35 (PDF pp. 35) · Review: machine-drafted-source-checked

services and supply agreements are the principal category without a substantial readability increase

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 35, that services and supply agreements are the principal category without a substantial readability increase. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because heterogeneity across categories can help future work identify causes rather than attributing all change to one universal force. It connects to services agreements, supply contracts, category heterogeneity, causal research, readability change, comparative trends.

printed pp. 35 (PDF pp. 35) · Review: machine-drafted-source-checked

consumer contracts did not become materially longer over the observed period

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 36, that consumer contracts did not become materially longer over the observed period. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because digitization and cheap reproduction did not generate the expected steady expansion in these forms. It connects to contract length, consumer contracts, digitization, copying costs, time trends, empirical surprise.

printed pp. 36 (PDF pp. 36) · Review: machine-drafted-source-checked

most commercial-contract categories also remained relatively stable in length

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 36, that most commercial-contract categories also remained relatively stable in length. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because greater semantic difficulty cannot be explained simply by adding more words or sentences. It connects to commercial contracts, length stability, semantic complexity, word counts, sentence counts, causal explanation.

printed pp. 36 (PDF pp. 36) · Review: machine-drafted-source-checked

year-to-year variation shows that contracts change despite extensive copying and reuse

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 36, that year-to-year variation shows that contracts change despite extensive copying and reuse. The discussion situates this proposition within III. Findings and the article's empirical reassessment of the plain-language movement. This is significant because boilerplate persistence coexists with ongoing addition, deletion, and revision of provisions. It connects to boilerplate, contract evolution, copy and paste, year variation, provision change, future research.

printed pp. 36 (PDF pp. 36) · Review: machine-drafted-source-checked

consumer contracts are no harder by these metrics than daily news, undermining decades of generalized complexity claims

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 37, that consumer contracts are no harder by these metrics than daily news, undermining decades of generalized complexity claims. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because consumer regulation should not target an abstract linguistic deficit that the benchmark does not reveal. It connects to consumer contracts, daily news, plain-language premise, benchmark evidence, regulatory rationale, contract complexity.

printed pp. 37 (PDF pp. 37) · Review: machine-drafted-source-checked

the absence of a broad linguistic crisis is not a rosy finding about consumer markets

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 37, that the absence of a broad linguistic crisis is not a rosy finding about consumer markets. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because harm may arise from market structure, constrained choice, substantive terms, or enforcement rather than difficult sentences. It connects to consumer harm, market structure, substantive regulation, choice, enforcement, linguistic complexity.

printed pp. 37 (PDF pp. 37) · Review: machine-drafted-source-checked

relatively readable consumer contracts are surprising under theories that firms benefit from obfuscation

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 37, that relatively readable consumer contracts are surprising under theories that firms benefit from obfuscation. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because the result points either to unexplored forces favoring clarity or to other mechanisms of firm advantage. It connects to firm obfuscation, market forces, consumer inattention, contract theory, empirical puzzle, research agenda.

printed pp. 37 (PDF pp. 37) · Review: machine-drafted-source-checked

post-sale rights vindication may be obstructed by structural or substantive causes other than readability

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 37, that post-sale rights vindication may be obstructed by structural or substantive causes other than readability. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because making forms easier does not necessarily improve dispute resolution, satisfaction, or enforcement. It connects to ex post use, rights vindication, dispute resolution, consumer satisfaction, structural barriers, contract enforcement.

printed pp. 37 (PDF pp. 37) · Review: machine-drafted-source-checked

Wittgenstein's ruler principle requires treating every measurement as evidence about the metric itself

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 38, that Wittgenstein's ruler principle requires treating every measurement as evidence about the metric itself. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because contract scores reveal weaknesses in readability formulas as well as features of the documents. It connects to Wittgenstein's ruler, measurement reflexivity, readability formulas, construct validity, contract data, methodology.

printed pp. 38 (PDF pp. 38) · Review: machine-drafted-source-checked

readability tests fail on reliability, inter-test coherence, and connection to consumer outcomes

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 38, that readability tests fail on reliability, inter-test coherence, and connection to consumer outcomes. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because the three defects jointly undermine claims of scientific precision. It connects to reliability, construct coherence, consumer outcomes, scientific merit, readability metrics, validity.

printed pp. 38 (PDF pp. 38) · Review: machine-drafted-source-checked

averaging across tests mitigates but cannot cure weak construct validity

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 38, that averaging across tests mitigates but cannot cure weak construct validity. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because a robust ensemble is useful for comparison without becoming a valid universal accessibility test. It connects to measurement ensemble, construct validity, robustness, universal metric, comparative analysis, epistemic limits.

printed pp. 38 (PDF pp. 38) · Review: machine-drafted-source-checked

plain-language advocates face a dilemma if they reject unfavorable results because the underlying tests are unreliable

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 38, that plain-language advocates face a dilemma if they reject unfavorable results because the underlying tests are unreliable. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because the same objection removes the scientific evidence on which metric-based reform was built. It connects to internal critique, plain-language evidence, measurement dilemma, burden of proof, regulatory science, readability.

printed pp. 38 (PDF pp. 38) · Review: machine-drafted-source-checked

a more sophisticated universal readability test would remain vulnerable to laboratory-field gaps and gaming

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 39, that a more sophisticated universal readability test would remain vulnerable to laboratory-field gaps and gaming. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because adding variables or machine learning does not solve incentives or external validity. It connects to advanced metrics, Goodhart's law, field validity, machine learning, metric gaming, readability.

printed pp. 39 (PDF pp. 39) · Review: machine-drafted-source-checked

universal readability measurement is conceptually flawed because readers differ across culture, cognition, education, and context

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 39, that universal readability measurement is conceptually flawed because readers differ across culture, cognition, education, and context. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because accessibility is relational between a text, task, and person rather than an intrinsic scalar property. It connects to reader diversity, relational readability, culture, cognition, context, universal measurement.

printed pp. 39 (PDF pp. 39) · Review: machine-drafted-source-checked

the adult illiteracy myth illustrates deeply nested citation chains that end without evidentiary support

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 39, that the adult illiteracy myth illustrates deeply nested citation chains that end without evidentiary support. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because legal scholarship can transform repetition into apparent fact unless researchers audit the terminal source. It connects to citation audit, leprechauns, legal scholarship, adult literacy, source chains, epistemic rigor.

printed pp. 39 (PDF pp. 39) · Review: machine-drafted-source-checked

laws calibrated five to seven grades below actual target readers may undermine consumer protection

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 39, that laws calibrated five to seven grades below actual target readers may undermine consumer protection. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because oversimplification can waste resources, patronize audiences, and crowd out more effective interventions. It connects to regulatory calibration, oversimplification, target audience, opportunity cost, consumer protection, grade mandates.

printed pp. 39 (PDF pp. 39) · Review: machine-drafted-source-checked

people who genuinely struggle with literacy need multimodal and structural support rather than shorter sentences alone

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 39, that people who genuinely struggle with literacy need multimodal and structural support rather than shorter sentences alone. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because visuals, audio, translation, education, and socioeconomic reform address heterogeneous barriers more directly. It connects to multimodal access, visual aids, audio, translation, structural reform, literacy vulnerability.

printed pp. 39 (PDF pp. 39) · Review: machine-drafted-source-checked

laboratory success does not establish that a readability metric will work in real transactions

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 40, that laboratory success does not establish that a readability metric will work in real transactions. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because policy should demand field validation against comprehension and behavior. It connects to laboratory-field gap, field validation, consumer transactions, behavioral outcomes, metric evaluation, evidence-based policy.

printed pp. 40 (PDF pp. 40) · Review: machine-drafted-source-checked

complex metrics create more dimensions that sophisticated drafters can game

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 40, that complex metrics create more dimensions that sophisticated drafters can game. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because technical sophistication can reduce transparency and incentive compatibility. It connects to complex metrics, gaming, drafter incentives, transparency, Goodhart's law, regulatory design.

printed pp. 40 (PDF pp. 40) · Review: machine-drafted-source-checked

a single readability number cannot capture diverse texts, transactions, and readers

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 40, that a single readability number cannot capture diverse texts, transactions, and readers. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because context-sensitive accessibility is incompatible with a universal one-size-fits-all compliance score. It connects to context sensitivity, consumer diversity, transaction type, compliance score, accessibility, measurement limits.

printed pp. 40 (PDF pp. 40) · Review: machine-drafted-source-checked

consumer-contract readability remained static while commercial-contract complexity rose

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 41, that consumer-contract readability remained static while commercial-contract complexity rose. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because the divergent trajectories create a causal puzzle that neither technology nor plain-language advocacy readily explains. It connects to divergent trends, consumer contracts, commercial contracts, causal puzzle, technology, plain-language laws.

printed pp. 41 (PDF pp. 41) · Review: machine-drafted-source-checked

consumer stasis is consistent with ineffectiveness, saturation, or irrelevance of plain-language laws

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 41, that consumer stasis is consistent with ineffectiveness, saturation, or irrelevance of plain-language laws. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because the same descriptive result can support distinct mechanisms that require separate causal testing. It connects to ineffectiveness, saturation, irrelevance, causal hypotheses, consumer readability, policy evaluation.

printed pp. 41 (PDF pp. 41) · Review: machine-drafted-source-checked

each leading interpretation of consumer stasis casts doubt on continued broad plain-language reform

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 41, that each leading interpretation of consumer stasis casts doubt on continued broad plain-language reform. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because the laws either failed, merely prevented deterioration, or addressed a problem already near its optimum. It connects to policy failure, status quo, counterfactual, plain-language reform, consumer contracts, causal uncertainty.

printed pp. 41 (PDF pp. 41) · Review: machine-drafted-source-checked

commercial complexity may reflect production costs and organizational practices but those accounts do not explain the upward trend

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 41, that commercial complexity may reflect production costs and organizational practices but those accounts do not explain the upward trend. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because cheaper editing technology makes rising difficulty especially puzzling. It connects to production process, organizational costs, Baumol's cost disease, contract complexity, legal technology, research puzzle.

printed pp. 41 (PDF pp. 41) · Review: machine-drafted-source-checked

greater commercial-contract complexity may increase adoption costs and slow contractual innovation

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 42, that greater commercial-contract complexity may increase adoption costs and slow contractual innovation. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because documents that are harder to understand and adapt can create market-wide spillovers even between sophisticated parties. It connects to adoption costs, contract innovation, commercial complexity, spillovers, sovereign debt, sophisticated parties.

printed pp. 42 (PDF pp. 42) · Review: machine-drafted-source-checked

franchise disclosures are especially troubling because franchisees may lack the sophistication and leverage of large firms

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 42, that franchise disclosures are especially troubling because franchisees may lack the sophistication and leverage of large firms. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because formal disclosure can fail when the mandated material itself is exceptionally difficult. It connects to franchise disclosure, power imbalance, small business, formal compliance, information access, consumer-business boundary.

printed pp. 42 (PDF pp. 42) · Review: machine-drafted-source-checked

the costs of nearly 800 plain-language laws include drafting, enforcement, compliance, and foregone reforms

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 42, that the costs of nearly 800 plain-language laws include drafting, enforcement, compliance, and foregone reforms. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because well-intentioned policy should be judged against realistic alternative uses of regulatory attention. It connects to opportunity cost, regulatory resources, compliance, enforcement, alternative reforms, policy evaluation.

printed pp. 42 (PDF pp. 42) · Review: machine-drafted-source-checked

the burden of proof should shift to proponents of new plain-language mandates

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 42, that the burden of proof should shift to proponents of new plain-language mandates. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because future proposals should identify a specific market, valid measure, reader baseline, and credible benefit. It connects to burden of proof, targeted mandates, valid measures, market segment, adult baseline, cost-benefit analysis.

printed pp. 42 (PDF pp. 42) · Review: machine-drafted-source-checked

firm-side readability mandates are not incentive compatible and can reward gaming

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 43, that firm-side readability mandates are not incentive compatible and can reward gaming. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because regulated drafters can satisfy a manipulable score without making information genuinely useful. It connects to incentive compatibility, firm-side duties, metric gaming, obfuscation, compliance, consumer information.

printed pp. 43 (PDF pp. 43) · Review: machine-drafted-source-checked

drafting for the average reader redirects attention from diverse readers at the tails

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 43, that drafting for the average reader redirects attention from diverse readers at the tails. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because consumer protection should optimize for actual vulnerability rather than modal performance. It connects to average reader, distribution tails, consumer vulnerability, targeted design, accessibility, distributional policy.

printed pp. 43 (PDF pp. 43) · Review: machine-drafted-source-checked

the assumption that one document must serve every reader is a fallacy

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 43, that the assumption that one document must serve every reader is a fallacy. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because multiple adaptive presentations can resolve conflicts among language, attention, visual, and format needs. It connects to single-document fallacy, adaptive presentation, language access, visual needs, attention, personalization.

printed pp. 43 (PDF pp. 43) · Review: machine-drafted-source-checked

AI-powered smart readers can simplify, personalize, interpret, and benchmark contracts for individual users

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 43, that AI-powered smart readers can simplify, personalize, interpret, and benchmark contracts for individual users. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because consumer-side tools can supply transformations that a static firm-authored form cannot. It connects to smart readers, simplification, personalization, interpretation, benchmarking, consumer empowerment.

printed pp. 43 (PDF pp. 43) · Review: machine-drafted-source-checked

smart readers can present the same agreement differently to teenagers, visual learners, summary-preferring users, or Spanish speakers

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 43, that smart readers can present the same agreement differently to teenagers, visual learners, summary-preferring users, or Spanish speakers. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because personalized comprehension is a concrete alternative to a mythical universal plain-text standard. It connects to multilingual access, visual presentation, age adaptation, summaries, personalized contracts, accessibility.

printed pp. 43 (PDF pp. 43) · Review: machine-drafted-source-checked

smart readers can reduce reliance on professional intermediaries and narrow access-to-justice gaps

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 43, that smart readers can reduce reliance on professional intermediaries and narrow access-to-justice gaps. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because on-demand contract analysis gives consumers usable help at the moment of decision or dispute. It connects to legal intermediaries, access to justice, on-demand analysis, consumer decisions, contract disputes, legal technology.

printed pp. 43 (PDF pp. 43) · Review: machine-drafted-source-checked

smart-reader policy must address accuracy, bias, privacy, security, and liability

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 44, that smart-reader policy must address accuracy, bias, privacy, security, and liability. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because consumer-side technology is not self-validating and requires governance proportionate to reliance. It connects to AI governance, accuracy, bias, privacy, liability, smart readers.

printed pp. 44 (PDF pp. 44) · Review: machine-drafted-source-checked

standards and certification can build trust in consumer-facing contract tools

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 44, that standards and certification can build trust in consumer-facing contract tools. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because validated performance and clear accountability are prerequisites for broad assimilation. It connects to certification, technical standards, consumer trust, AI applications, accountability, legal frameworks.

printed pp. 44 (PDF pp. 44) · Review: machine-drafted-source-checked

challenging plain-language metrics is not a license for firms to obfuscate

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 44, that challenging plain-language metrics is not a license for firms to obfuscate. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because the failure of one reform paradigm creates a duty to search for better interventions, not to accept harm. It connects to anti-obfuscation, policy innovation, consumer harm, plain language, regulatory alternatives, firm conduct.

printed pp. 44 (PDF pp. 44) · Review: machine-drafted-source-checked

substantive regulation may be necessary when unfair contract terms cause the underlying harm

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 44, that substantive regulation may be necessary when unfair contract terms cause the underlying harm. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because linguistic form should not shield policymakers from evaluating price, rights, and fairness. It connects to substantive regulation, unfair terms, contract content, consumer harm, market fairness, regulatory focus.

printed pp. 44 (PDF pp. 44) · Review: machine-drafted-source-checked

comparison tools, fairness ratings, and AI shopping assistants can strengthen consumer choice

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 44, that comparison tools, fairness ratings, and AI shopping assistants can strengthen consumer choice. The discussion situates this proposition within IV. Interpretation and Normative Implications and the article's empirical reassessment of the plain-language movement. This is significant because market and technological interventions may address sophistication gaps more directly than prose mandates. It connects to comparison tools, fairness ratings, AI shopping assistants, market competition, consumer choice, sophistication gap.

printed pp. 44 (PDF pp. 44) · Review: machine-drafted-source-checked

the plain-language movement achieved immense legal success while failing scrutiny on each foundational empirical premise

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 45, that the plain-language movement achieved immense legal success while failing scrutiny on each foundational empirical premise. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because policy influence and policy effectiveness must be evaluated separately. It connects to movement success, empirical failure, plain language, policy effectiveness, consumer revolution, regulatory evaluation.

printed pp. 45 (PDF pp. 45) · Review: machine-drafted-source-checked

simplification is not a silver bullet and appears to produce at most modest consumer-outcome effects

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 45, that simplification is not a silver bullet and appears to produce at most modest consumer-outcome effects. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because clearer text can be desirable without serving as a complete theory of consumer protection. It connects to simplification, consumer outcomes, silver bullet, plain drafting, policy limits, contract access.

printed pp. 45 (PDF pp. 45) · Review: machine-drafted-source-checked

readability is neither necessary nor sufficient to cure market pathology

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 45, that readability is neither necessary nor sufficient to cure market pathology. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because fair standard terms may work despite imperfect prose, while an exploitative bargain remains harmful even when disclosed beautifully. It connects to necessary and sufficient conditions, market pathology, substantive fairness, disclosure, contract terms, consumer welfare.

printed pp. 45 (PDF pp. 45) · Review: machine-drafted-source-checked

a clearly disclosed payday loan with a 600 percent APR remains a rotten deal

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 45, that a clearly disclosed payday loan with a 600 percent APR remains a rotten deal. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because the quality of the consumer experience and form cannot substitute for evaluation of the transaction's substance. It connects to payday lending, APR, clear disclosure, substantive harm, consumer experience, market regulation.

printed pp. 45 (PDF pp. 45) · Review: machine-drafted-source-checked

consumer law should focus more directly on market structure, choice, and substantive terms

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 45, that consumer law should focus more directly on market structure, choice, and substantive terms. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because these variables better explain serious harm than the generalized claim that consumers cannot read. It connects to market structure, consumer choice, substantive terms, regulatory priorities, contract doctrine, consumer protection.

printed pp. 45 (PDF pp. 45) · Review: machine-drafted-source-checked

consumer-assistive AI and contract APIs can serve people with individualized accessibility needs

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 45, that consumer-assistive AI and contract APIs can serve people with individualized accessibility needs. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because regulators can enable user-controlled interaction with agreements instead of relying on firms' goodwill. It connects to consumer-assistive AI, APIs, smart readers, accessibility, regulatory design, contract interaction.

printed pp. 45 (PDF pp. 45) · Review: machine-drafted-source-checked

legal movements can perpetuate myths when noble ends substitute for empirical underwriting

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 45, that legal movements can perpetuate myths when noble ends substitute for empirical underwriting. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because good intentions do not excuse weak evidence after decades of costly reform. It connects to legal movements, myth propagation, good intentions, empirical accountability, reform costs, legal scholarship.

printed pp. 45 (PDF pp. 45) · Review: machine-drafted-source-checked

the relevant retrospective question is whether half a century of plain-language reform caused consumers to read, understand, and decide better

Professor Yonathan Arbel claims, in the article “The Readability of Contracts: Big Data Analysis” on page 46, that the relevant retrospective question is whether half a century of plain-language reform caused consumers to read, understand, and decide better. The discussion situates this proposition within Conclusion and the article's empirical reassessment of the plain-language movement. This is significant because policy should ultimately be evaluated by human outcomes rather than enactment counts or formula compliance. It connects to retrospective evaluation, consumer understanding, informed decisions, policy outcomes, plain language, empirical accountability.

printed pp. 46 (PDF pp. 46) · Review: machine-drafted-source-checked

Machine Files

Full Text Entry Point

The cleaned full text is exposed at fulltext_clean.txt, with fulltext_raw.txt preserved for audit. The compatibility path fulltext.txt points to the cleaned text. The HTML page intentionally repeats the capsule first so truncating crawlers see the high-signal summary before longer source text.