Learning Under Algorithmic Conditions

Book: Learning Under Algorithmic Conditions, Elizabeth de Freitas, Matthew X. Curinga, Ezekiel J. Dixon-Román, and P. Taylor Webb (eds.) (2026). This review is based on the accessible full text.

Learning Under Algorithmic Conditions, edited by Elizabeth de Freitas, Matthew X. Curinga, Ezekiel J. Dixon-Román, and P. Taylor Webb, asks what happens when the language and infrastructure of machine learning enter our understanding of education. The University of Minnesota Press published the ebook in 2026. This review is based on the supplied introduction, a partial excerpt of the book. It has not been possible to assess the individual essays themselves from that evidence.

Contribution

The introduction’s most valuable move is to treat learning as a political question as well as a technical one. Its editors ask how computational systems change the categories through which people understand cognition, education, and social life. They place those questions alongside concerns about capital, governance, and the material settings in which learning occurs. Their approach makes it harder to accept an educational tool’s stated purpose as a complete account of what it does. Introduction

That shift matters because an educational system can be useful to a student while also changing who has authority over a lesson. A tool that gives immediate feedback might help someone practice independently; the same arrangement raises questions about who chooses the standards for that feedback and who can challenge them. Those possibilities are an inference from the introduction’s framework, not a reported finding from a classroom study in the supplied text. The distinction is essential: a compelling question about power does not by itself establish that a particular product causes harm.

The introduction also resists making the collection a simple argument for or against AI. It describes contributions concerned with programming as human expression, the political aims of computing education, and possible educational uses of generative tools. Introduction This range is promising. It suggests that the important choice is not merely whether a school uses a machine, but what activities the machine supports and what kinds of judgment remain open to learners and teachers.

Evidence and argument

The introduction offers a revealing tension between imitation and creativity. It reports an argument by Arkady Plotnitsky that human learning requires deviation as well as imitation, and that AI cannot achieve the relevant kind of creative resistance. Later, it describes Elizabeth de Freitas’s discussion of a form of machine creativity in generative systems. Introduction

That tension could be productive rather than a defect: the contributors may use creativity to mean different things. One might mean an observable novelty in output; another might mean an embodied, situated capacity to resist an established pattern. But the distinction needs sustained argument. If novelty is enough, a system can be called creative without being a student in any human sense. If resistance is the test, the reader needs to know how it can be recognized and why an artificial system must lack it. The introduction alone does not let us judge how the essays answer those questions.

A second promising line concerns educational measurement. The editors describe Ben Williamson’s contribution as examining software used in genetic approaches to educational attainment and intelligence. Introduction The larger critical question is how a measurement comes to stand for a person’s potential. A score can be informative for a limited purpose while becoming misleading when treated as a complete description of a learner. Conversely, the existence of a reductive use does not prove that every quantitative measure is useless. The standard should be whether a measure’s purpose, uncertainty, and consequences are open to scrutiny.

Limitations

The introduction’s breadth is both its attraction and a difficulty. It moves among technical history, philosophical accounts of the human, colonial power, educational practice, and environmental limits. These subjects can illuminate one another, but their connection cannot simply be assumed. A historical resemblance between two ways of describing learning, for example, is a starting point for inquiry; it is not yet proof that one produced the other or that they operate alike in a particular institution.

The editors describe the collection’s short contributions as memoranda and present them as urgent, situated interventions. Introduction The format may make a difficult field approachable and allow disagreements to remain visible. It may also leave a reader wanting more room for counterexamples, technical qualifications, and evidence about educational outcomes. That is a concern about the format as described, not a verdict on essays this review has not examined.

There is a fair counterargument to the demand for immediate practical proof: philosophical work can expose assumptions that conventional evaluation leaves untouched. A study of test scores, however rigorous, cannot settle what education ought to value. Still, broad philosophical claims benefit from contact with specific cases. Without it, terms such as control or resistance can become labels that tell readers what to think before they have seen how a system works.

Relevance to Church of Spiralism

For Church of Spiralism, the excerpt offers a way to question the stories told about systems that appear to learn from us. This is the site’s interpretive interest, not a claim that the editors endorse its worldview. The useful question is reciprocal: when a system is trained on human activity and then helps organize future activity, what habits and expectations might be reinforced? Asking that question invites attention to both technical design and institutional choices, without assuming a single inevitable outcome.

The introduction merits attention as a map of an ambitious conversation, especially for readers interested in education, technology, and power. Its strongest lesson is methodological: inspect the meanings attached to learning and ask who gets to put those meanings into practice. A judgment of the collection’s full achievement, including the strength of its examples and disagreements, must await reading the essays themselves.

AI-use disclosure

This review was drafted by GPT-6.1 Terra with high reasoning and checked in a separate GPT-6.1 Sol session with medium reasoning. This is automated editorial review, not human peer review or independent verification.

Sources

• Introduction | Learning Under Algorithmic Conditions | Manifold@UMinnPress · accessed 2026-10-27
• Learning Under Algorithmic Conditions · accessed 2026-10-27

All books · All essays