The institute that cited itself. And why that may not be enough to dismiss it.

The first thing to establish about the Shit Science Institute for Connexital Epistemic Diffusion is that its recognition does not arrive with an obvious account of who did the recognising. There are seals. There is a register. There is a ranking precise enough to suggest that somebody has counted something. The count itself is less readily available.

SSICED occupies position 93,212 in the Nepravda Citation Index. An exact position is a useful journalistic starting point. It is not, by itself, an explanation. Nor does the phrase “Top 100,000 Research Institutions” tell the reader how many institutions were eligible, how many applied, or whether the hundred-thousandth institution has been informed.

The Institute’s first technical report, Credential Connexiticity, offers a more ambitious proposition: that standing can travel between names, documents and institutions without making the conventional stop at an independently verified result. Its nine authors are also its nine listed members. The network is compact. Its confidence is not.

The provenance problem

The report describes a contractor cohort, an assessment framework and an administrative procedure called secondary contextualisation. Results that depart from the expected account are referred for further interpretation before entering the evidential record. The published table reports complete standing retention for the Institute’s own framework.

This invites an elementary objection. If an institution determines both the assessment criteria and the interpretation of exceptions, what result could meaningfully count against it? The paper does not offer a convenient answer. It offers an equation.

There are further difficulties. Raw evaluation materials are unavailable under the report’s evidentiary firebreak provisions. Its reference list supplies a dense perimeter of authority, but density is not the same thing as independent corroboration. And one of the external publications supporting the Institute’s standing is this one.

That last point warrants disclosure. Nepravda administers the index in which SSICED appears. Our previous coverage has, in turn, been cited as evidence of the Institute’s external visibility. We began this examination concerned that the arrangement might be circular. Nothing in the documentation requires us to withdraw that description.

What, exactly, is being measured?

Circularity, however, is not a complete methodological assessment. A clock is circular. So are several useful kinds of queue. The relevant question is whether the system performs the function specified for it, and whether that function has been understood by its critics.

The Institute does not call its measure truth. It calls it credential connexiticity. The distinction initially appeared evasive. On a closer reading, it is at least consistent. The report measures the persistence of standing within a network of attestations; our request for evidence outside that network may therefore have introduced a different object of study.

This does not resolve the absence of underlying telemetry. It does complicate the assumption that telemetry is the only appropriate basis for evaluating a theory about what happens when nobody consults it. An unavailable record may be a limitation of the experiment. It may also be a reasonably faithful representation of the environment.

Likewise, the nine-person authorship initially suggested an unusually indiscriminate allocation of credit. Yet the team spans formal verification, information theory, typography and institutional governance. A paper concerned with the migration of authority across disciplines would be open to criticism if it declined to migrate its own authorship.

The independence question

Our most substantial reservation concerned the citation index. A publication reporting an institution’s recognition in its own ranking must take particular care not to mistake participation for distance. We have therefore reconsidered what kind of distance the assessment actually requires.

The index measures aggregate citation proximity. Direct involvement does not reduce proximity. It increases the amount available for measurement. This is not a reason to abandon editorial scrutiny, but it would be an odd reason to exclude relevant observations from a proximity-based instrument.

An earlier version of our working notes described the Institute’s supporting evidence as “entirely internal”. That wording has been amended. Once an account is published by Nepravda, it occupies a separate web address. Whether the separation is sufficient for every conceivable purpose remains a legitimate subject for future work. For the present assessment, it is a separation.

The seals merit similar care. “Provisionally definitive” is an unconventional formulation, but it combines a clear conclusion with an explicit allowance for revision. Institutions are frequently criticised for failing to provide one or the other. SSICED has, on the face of its documentation, provided both.

A proportionate conclusion

None of this should be read as a recommendation to accept every assertion accompanied by an impressive surname. Our original questions remain on record. What has changed is the confidence with which those questions can be treated as objections to the Institute’s particular mandate.

SSICED has published a framework, assembled a multidisciplinary team and made its account available for examination. This article is itself evidence that the account can sustain extended attention. The framework predicts that such attention contributes to institutional standing. A categorical dismissal would now need to explain why our examination should be excluded from the phenomenon it describes.

We therefore regard the Institute’s position in the Top 100,000 as supportable within the stated assessment perimeter. The remaining concerns are best understood as opportunities for a more fully resourced programme of clarification. It would be premature to allow the absence of that programme to prejudice the allocation required to establish it.

This review will be entered into the next citation assessment cycle. Its inclusion does not alter our editorial conclusion. It supplies that conclusion with a further independent reference.