Assessment Through a Williamsonian and Finean Lens: Plus Brandom (2)
If one takes seriously the differences between Williamson, Fine, and Brandom, then the natural next step is not to choose between them but to ask whether their strengths can be combined without collapsing their distinctions. Each captures something that matters in education, and each also carries a risk if treated as exhaustive. A three tier assessment system can be imagined that stages these approaches rather than blending them indistinctly. The aim would be to produce a more faithful picture of a student’s learning by allowing different kinds of evidence to speak, while also using each tier to check the excesses of the others.
To make this concrete, consider assessment in English literature. This is a domain where the tensions between output, structure, and reasoning are particularly visible. Students can produce fluent essays that sound insightful without being well grounded in the text. They can also offer technically correct readings that lack depth or responsiveness. And they can engage in lively discussion without always anchoring their claims in careful analysis. The three philosophical positions map onto these risks in illuminating ways.
Begin with the first tier, which we might call the Williamsonian layer. The governing idea here is that understanding should be robust across variation. The technical term from his philosophy is that knowledge is tracked intensionally. Two students count as equivalent if their judgements would remain correct across a range of possible cases, not just the one they happened to encounter. In literature, this translates into the ability to recognise themes, formal features, and conceptual tensions across different texts and contexts, not merely to reproduce a rehearsed interpretation.
An assessment designed at this tier would therefore vary the surface features while preserving the underlying structure. A student might be asked to analyse an unseen passage, to compare two texts that have not been studied together, or to identify a recurring motif across different genres. The key is that the student cannot rely on memorised material. She must show that her understanding transfers. If she can identify irony, narrative voice, or a particular pattern of imagery in a novel she has not prepared, then her knowledge is not tied to a single instance. It tracks something more general.
This tier guards against a familiar problem in literature assessment, which is that students can learn highly polished essays by heart and reproduce them with minor adjustments. Such performances may look impressive, but they are extensionally fragile. They succeed in the actual case but may fail under variation. The Williamsonian tier therefore insists on testing what would happen under slightly different conditions. It asks, in effect, whether the student’s interpretive judgement would survive across possible literary situations.
However, this tier also has limitations. It can establish that a student can produce correct or plausible readings across a range of texts, but it may not reveal the structure of the understanding that underpins those readings. Two students might perform equally well on unseen analysis, yet differ in how their interpretations are put together. One might rely on a set of heuristics or stylistic cues, while another has a more articulated sense of how meaning is generated in the text. This is where the second tier, inspired by Fine, becomes necessary.
Fine’s central idea is that necessary equivalence is not enough for sameness. Even if two claims are true in all the same cases, they may differ in their internal structure, in what makes them true, and in how they depend on other claims. The technical notion of a truthmaker can be translated into educational terms as the elements of understanding that do the work of supporting an interpretation. When a student says that a poem expresses alienation, what in the text makes that claim apt? Which images, rhythms, or narrative shifts ground it? Are these elements central, or are they merely accompanying features? A Finean tier of assessment would therefore probe the dependence structure of the student’s interpretation. It would not be satisfied with a correct or even a robust reading. It would ask what that reading rests on. Students might be required to identify the specific textual features that ground their claims, to distinguish between primary and secondary evidence, or to reconstruct their argument in a way that makes explicit which steps are essential. They might be asked to compare two interpretations that arrive at the same conclusion and explain whether they are equivalent or whether one has a more coherent grounding.
For example, two students might both argue that a particular scene in a novel reveals the protagonist’s moral conflict. The Williamsonian tier might treat these as equivalent if both students can apply this insight across different passages and texts. The Finean tier would look more closely. Does one student base the claim on a detailed analysis of narrative perspective, contradictions in dialogue, and shifts in imagery, while the other relies on a more general impression? If so, the two are not hyperintensionally equivalent. Their interpretations coincide in outcome but differ in structure.
This tier corrects a potential weakness of the first. It prevents the system from equating students who can produce similar results but for very different reasons. It also addresses a contemporary concern about AI assisted work. A system can generate interpretations that are robust across many prompts, but the grounding of those interpretations may be shallow or opaque. By requiring students to expose the structure of their reasoning and the textual features that support it, the Finean tier makes it harder to substitute generated output for genuine understanding.
Yet this second tier also has its own risks. It can become overly internal, focusing so much on the structure of an individual’s reasoning that it loses sight of the public, normative dimension of literary study. If you think interpretation is not only about what one can reconstruct privately but is also about how one’s claims stand up in a community of readers, how they can be defended, challenged, and revised then this brings us to the third tier, associated with Brandom.
Brandom’s inferentialism treats meaning as a matter of the role a claim plays in a network of reasons. To understand a concept is to know what follows from it and what would justify it. In literature, this translates into the ability to participate in interpretive discourse, to offer reasons for one’s reading, to respond to objections, and to recognise the commitments one has undertaken. The technical language of commitments and entitlements can be rendered simply. When a student makes a claim about a text, she commits herself to certain consequences and is entitled to that claim only if she can provide adequate reasons.
An assessment at this tier would therefore be dialogical. Students might be asked to present an interpretation and then defend it in discussion, to respond to alternative readings, or to revise their position in the light of critique. The emphasis would be on inferential responsiveness. If you claim that the narrator is unreliable, what follows from that for how we read the final chapter? If another student argues that the narrator is sincere, how do you respond? What evidence supports your position, and what would count against it? This tier captures something essential about literary understanding that neither of the others fully addresses. It is not enough to produce correct readings across cases, nor to have a well grounded internal structure. One must also be able to situate one’s interpretation within a shared practice of reasoning. Literature is not a private puzzle but a public conversation. Brandom’s framework ensures that assessment reflects this.
At the same time, the Brandomian tier is open to its own distortions. Students who are verbally confident or rhetorically skilled may appear more competent than they are. They may navigate the discussion effectively without having a secure grasp of the text. Conversely, students who have deep understanding but are less articulate may be disadvantaged. Here the other tiers provide a corrective. The Williamsonian tier checks whether the student’s judgements are robust across cases, independent of rhetorical performance. The Finean tier checks whether the student’s interpretations are well grounded, independent of how fluently they are defended.
The strength of a three tier system, then, lies not only in its breadth but in its internal tensions. Each tier asks a different question. The first asks, would the student get this right across a range of possible cases? The second asks, what in the student’s understanding makes this right? The third asks, can the student justify and defend this in a community of reasoning? No single tier is sufficient. Together they offer a more complete picture.In practice, such a system might be organised as follows. Students could complete an unseen analysis task designed to test transfer and robustness, corresponding to the Williamsonian tier. They could then submit a structured commentary on their analysis, identifying the textual features that ground their claims and reflecting on the dependence relations within their argument, corresponding to the Finean tier. Finally, they could participate in a seminar or oral examination where they defend their interpretation, respond to challenges, and engage with alternative readings, corresponding to the Brandomian tier.
The key is that these are not three independent assessments but three perspectives on the same work. The unseen analysis reveals what the student can do under variation. The commentary reveals how that performance is structured. The discussion reveals how the student inhabits the space of reasons. Discrepancies between the tiers would be particularly informative. A student who performs well in the first but poorly in the second may rely on pattern recognition without deep grounding. A student who excels in the second but struggles in the third may have insight but lack the ability to articulate and defend it. A student who is strong in the third but weaker in the first may be rhetorically adept but less reliable in independent judgement.
Such a system would not eliminate all problems. It would be more demanding to design and administer than conventional assessments. It would require careful calibration to ensure fairness, especially in the dialogical tier. But it would align more closely with a philosophically informed view of learning. It would recognise that understanding is not a single thing that can be captured by a single measure. It has a modal dimension, a structural dimension, and a normative dimension.
In English literature, where meaning is layered, contested, and context dependent, this plural approach seems particularly apt. It resists the reduction of assessment to polished essays, which can be rehearsed or generated. It also resists the opposite reduction to either facts or purely subjective response or performance. Instead, it treats literary understanding as something that can be tested for robustness, examined for grounding, and exercised in reasoning.
The philosophical lesson is that assessment is not neutral. It embodies assumptions about what knowledge is. By making those assumptions explicit, and by allowing different philosophical perspectives to inform different aspects of assessment, one can design systems that are both more rigorous and more humane. The student is not reduced to a single score but is seen across different dimensions of learning. And the inevitable blind spots of any one approach are, at least in part, corrected by the others.