The Pragmatic Maxim
The pragmatic maxim, in Peirce's mature formulation, holds that to understand what a concept means, consider what effects the objects falling under that concept would have in the full range of conceivable practical situations, and the sum of those effects exhausts the concept's meaning. The maxim is not a theory of truth. It is a method of clarification — a tool for stripping away verbal confusion and revealing whether a concept that seems to say something actually says anything at all, or merely produces a warm feeling of comprehension without determinate content. Peirce later renamed his doctrine pragmaticism — "ugly enough to be safe from kidnappers" — to distinguish it from William James's looser versions. Pragmaticism is not the doctrine that ideas are valuable insofar as useful; it is the doctrine that meaning is constituted by practical consequences, and that concepts specifying no determinate consequences are, however eloquent, meaningless.
The Institutional Capture Problem — Contrarian ^ Opus
There is a parallel reading of the pragmatic maxim's application to AI discourse that begins from the political economy of knowledge production rather than philosophical method. While Peirce's maxim promises to strip away meaningless concepts by demanding specification of practical consequences, the actual deployment of such diagnostic tools in contemporary AI debates occurs within institutions already captured by the very interests producing the conceptual confusion. The companies developing AI systems control both the discourse about their meaning and the metrics by which their practical consequences are measured. When OpenAI or Anthropic define what counts as "alignment" or "safety," they simultaneously create the concepts and specify which practical consequences matter. The pragmatic maxim, in this context, becomes not a neutral diagnostic but a tool whose application is predetermined by who gets to define "practical" and "consequence."
The deeper problem is that the maxim assumes a community of inquiry capable of collectively determining practical consequences — Peirce's community of investigators converging on truth through shared method. But AI development occurs within proprietary black boxes, where the actual practical consequences (computational costs, environmental impacts, labor displacement patterns) are deliberately obscured while speculative consequences ("superintelligence," "human flourishing") dominate discourse. The corporations shaping AI can make any concept appear to have determinate practical consequences simply by building systems that produce those consequences, regardless of whether the concepts actually clarify or obscure. The pragmatic maxim thus faces a bootstrapping problem: it requires access to practical consequences to evaluate concepts, but the entities controlling those consequences also control the concepts. In this reading, conceptual confusion in AI discourse isn't a bug but a feature — maintaining ambiguity about what AI "means" allows maximum flexibility in reshaping labor relations, extracting value, and avoiding regulation.
In the AI Story
The maxim is Peirce's diagnostic instrument for conceptual analysis. Applied to AI discourse, it strips away concepts that sound meaningful but fail to specify testable practical consequences. The Peirce volume uses the maxim to test the concept of amplification in The Orange Pill — and finds that the practical consequences the metaphor specifies do not match the observed phenomena. The concept does not survive pragmaticist scrutiny.
The maxim's power lies in its refusal to treat sounding-meaningful as equivalent to being-meaningful. Many claims in contemporary AI discourse have the grammatical form of substantive propositions without specifying any determinate consequences that could confirm or disconfirm them. "AI will transform everything" has this character — it sounds like a claim, but absent specification of which practical consequences would constitute transformation, it cannot be evaluated.
Peirce's distinction between pragmatism and pragmaticism matters. Pragmatism, in James's version, became a theory of truth (true ideas are those that work). Pragmaticism, in Peirce's strict version, is a theory of meaning (concepts mean what they would do in all conceivable practical situations). The distinction preserves the realist commitment that some ideas track reality and others do not — a commitment James's pragmatism tended to erode.
The maxim provides, at the level of concept analysis, what Secondness provides at the level of experience: a test against brute consequence. A concept whose practical consequences cannot be specified is a concept floating free of the resistance of reality — symbol without index, to use the semeiotic vocabulary — and is in the same relation to genuine meaning as the AI's hall of mirrors is to genuine understanding.
Origin
Peirce first articulated the maxim in "How to Make Our Ideas Clear" (1878), the second essay in the Illustrations of the Logic of Science series.
He renamed his doctrine pragmaticism in 1905 after James's popularization of pragmatism had, in Peirce's view, dissolved the specific rigor of the original maxim.
Key Ideas
Meaning, not truth. A method of clarifying what concepts mean, not a theory of which beliefs are true.
Practical consequences as criterion. The meaning of a concept consists in the effects its objects would have in all conceivable practical situations.
Diagnostic against verbal confusion. Concepts that specify no determinate consequences are, however eloquent, meaningless.
Pragmaticism, not pragmatism. Peirce's strict version preserves realism against James's looser interpretation.
Appears in the Orange Pill Cycle
Scales of Pragmatic Application — Arbitrator ^ Opus
The weight between these views shifts dramatically depending on which scale and timeframe we examine. At the level of immediate technical discourse — when engineers specify what a particular model can do — the pragmatic maxim works essentially as Peirce intended (90% original view). A claim about GPT-4's capabilities can be tested against determinate consequences; the concept of "token prediction" has clear practical meaning. The maxim successfully distinguishes genuine technical concepts from marketing vapor. But move to the level of institutional discourse about AI's societal implications, and the contrarian view gains force (70% contrarian). Here, the entities defining concepts also control the conditions under which practical consequences manifest. "AI safety" means what the labs building AI systems need it to mean.
The synthesis emerges when we recognize that the pragmatic maxim operates differently at different scales of social organization. In small communities of practice with shared access to phenomena, it functions as Peirce envisioned — a collective tool for clarifying meaning through experimental consequence. In large-scale corporate-dominated discourse, it becomes subject to capture, with practical consequences themselves manufactured to validate predetermined concepts. The maxim retains its philosophical validity while its social application becomes politically contested.
The frame that holds both views is temporal: the pragmatic maxim is both a present diagnostic tool and a future-oriented regulative ideal. Right now, its application to AI discourse is compromised by asymmetric power over what counts as consequence. But it also provides the standard by which that compromise becomes visible. The maxim's demand for determinate practical consequences creates pressure toward eventual clarity — even manufactured consequences eventually collide with unmanufactured reality. The question isn't whether the maxim works, but how long the gap between conceptual confusion and practical clarity can be maintained, and who bears the cost of that gap.
Further reading
- Charles Sanders Peirce, "How to Make Our Ideas Clear" (1878)
- Charles Sanders Peirce, "What Pragmatism Is" (1905)
- Christopher Hookway, Peirce: The Arguments of the Philosophers (Routledge, 1985)
- Cheryl Misak, The American Pragmatists (Oxford, 2013)
- Robert Talisse and Scott Aikin, Pragmatism: A Guide for the Perplexed (Continuum, 2008)