Too new to judge.

An impressive demonstration can be genuine while leaving its central claim unresolved. Research on radical innovation asks what an audience needs to make a warranted judgment, and why a better explanation can lead to informed rejection.

A warehouse robot completes a polished demonstration. The recording is genuine. The movement is convincing. A buyer still needs to decide what the demonstration establishes. Does it show autonomous response to changing conditions, or could a scripted sequence produce the same visible result?

That distinction is the starting point of my conceptual manuscript, Too New to Judge: Evidential Relevance and the Evaluability of Radical Innovation. It asks how a recipient becomes able to evaluate an unfamiliar offering under a stated standard. The question concerns the reasons supporting a judgment, rather than the enthusiasm the offering generates.

A genuine display leaves several possible judgments

The robot example is a constructed comparison, with a fixed recording and reference demonstration. Under an appearance or styling standard, the visible result can support a favourable judgment whether the behaviour was autonomous or scripted. Under an autonomy standard, the difference between those processes becomes decisive.

A parallel performance example separates visible composition from live responsiveness. An audience can judge the composition while remaining uncertain whether digital performers responded in real time or followed playback. The same recording supplies enough information for one comparison and too little for another.

The research calls this criterion-conditioned evaluability. To complete a warranted comparison, the evaluator needs an explicit criterion, a comparator, relevant evidence and an inference another person could reconstruct. A positive, negative or tied verdict can all meet that requirement. Recognising that important evidence is missing is a justified deferral, which remains a different outcome.

Evidence matters through the question it can settle

A production detail is relevant when changing it could change the verdict under the required standard. Whether the robot used a script matters directly to an autonomy claim. It may leave a stipulated styling comparison untouched. What the audience must know therefore depends on what it has been asked to judge.

The manuscript distinguishes authentication, feasibility and attribution. Authentication asks whether the specified recording claim is genuine. Feasibility asks whether a proposed process could produce the outcome. Attribution asks which process actually did. Several processes can be feasible even though only one accounts for the demonstration.

A familiar category label can help people recognise a comparison family without supplying a usable standard or identifying the production process. A compelling narrative can similarly improve reception while leaving the decisive factual contrast unresolved. The contribution is to specify what additional content would turn those aids into reasons for judgment.

When understanding produces an unfavourable answer

Suppose a valid, conclusive test establishes that the robot’s demonstration was scripted. Belief in the advertised autonomous process falls. At the same time, the autonomy comparison becomes complete: the evaluator now has the evidence needed to reach an unfavourable verdict. The earlier styling judgment can remain favourable.

This is an important distinction for innovation management. A communication effort can make a claim more evaluable while making it less commercially attractive. Treating every negative reaction as failed explanation can encourage further persuasion when the appropriate response is to qualify the claim, improve the product or acknowledge a limitation.

Conclusive resolution matters. Evidence that merely makes scripting more likely may still leave both explanations possible. The recipient can become more doubtful without completing the comparison. The paper keeps declining belief, warranted rejection and warranted deferral separate so that one attitude measure cannot stand in for all three.

Changing the standard changes the task

Now consider a different event. Before the production test, the buyer replaces a styling requirement with an autonomy requirement. The recording and available evidence are unchanged. A comparison that was complete under the first standard no longer answers the commissioning question.

The manuscript describes this as displacement. The earlier answer remains warranted for its original task; a new required comparison is unresolved. This is distinct from losing information. It also differs from discovering that an earlier evidence set omitted a feasible alternative explanation.

Holding the standard and comparator fixed clarifies the logic. Valid new information can eliminate possibilities without overturning a verdict already shared by every remaining possibility. Whether a real recipient can retain, understand and use the supporting reason is an additional behavioural question. The conceptual account does not supply attention or competence by assumption disguised as an observation.

Two explanations can be complements or alternatives

The paper also examines how production evidence and clarification work together. In one task, a technical trace is useless without the rule connecting it to the criterion, while the rule is useless without the trace. Both are required to complete the inference.

In another task, either component can independently establish that a necessary requirement fails. The second explanation then adds another sufficient reason for the same verdict. Its smaller incremental contribution is redundancy in justification, rather than evidence that more information has damaged evaluability.

This distinction affects research design. Tasks must be classified by their declared evidence and inferential requirements before observing participant responses. Sorting them afterward using the same outcome cells that are supposed to test the interaction would build the conclusion into the classification.

Turn the claim into a testable account

The manuscript develops stipulated examples and logical restrictions, with an optional prospective study protocol. It reports no participant study, treatment effects or validated managerial instrument. The prevalence of the different task types and people’s ability to use sufficient reasons remain questions for empirical research.

A useful test would record the standard actually applied, the named comparator, the factual contrast capable of changing the verdict and the recipient’s criterion-linked reason. It would distinguish those observations from liking, confidence, adoption and belief in the advertised process. A favourable response alone cannot establish the proposed mechanism.

The interdisciplinary connection is between innovation research, epistemology, psychology and evaluation design. Novel offerings can challenge both what people believe happened and the standards by which they compare it. Making those two problems explicit gives producers and evaluators a more constructive conversation.

The practical ambition is to replace a vague demand for a more convincing demonstration with a precise question: which comparison remains unresolved, and what fact or explanation would settle it? That approach makes room for justified approval, informed rejection and honest deferral. Each can be a successful outcome of clearer judgment.