Symmetry’s mandate: Constraining the politicization of American administrative law

Daniel E. Walters

Research output: Contribution to journalArticlepeer-review


Recent years have seen the rise of pointed and influential critiques of deference doctrines in administrative law. What many of these critiques have in common is a view that judges, not agencies, should resolve interpretive disputes over the meaning of statutes-disputes the critics take to be purely legal and almost always resolvable using lawyerly tools of statutory construction. In this Article, I take these critiques, and the relatively formalist assumptions behind them, seriously and show that the critics have not acknowledged or advocated the full reform vision implied by their theoretical premises. Specifically, critics have extended their critique of judicial abdication only to what I call Type I statutory errors (that is, agency interpretations that regulate more conduct than the best reading of the statute would allow the agency to regulate) and do not appear to accept or anticipate that their theory of interpretation would also extend to what I call Type II statutory errors (that is, agency failures to regulate as much conduct as the best reading of the statute would require). As a consequence, critics have been more than willing to entertain an end to Chevron deference, an administrative law doctrine that is mostly invoked to justify Type I error, but have not shown any interest in adjusting administrative law doctrine to remedy agencies’ commission of Type II error. The result is a vision of administrative law’s future that is precariously slanted against legislative and regulatory action.

Original languageEnglish (US)
Article number2
Pages (from-to)455-518
Number of pages64
JournalMichigan Law Review
Issue number3
StatePublished - Dec 2020

All Science Journal Classification (ASJC) codes

  • Law


Dive into the research topics of 'Symmetry’s mandate: Constraining the politicization of American administrative law'. Together they form a unique fingerprint.

Cite this