Cross-Checking Repertory Rubrics with Remedy Provings: A Method for Validation

What the Repertory Actually Records

A repertory is an index, not a proving. Each rubric lists symptoms and points to remedies that were said to produce them. The remedy names were placed there by a compiler reading provings, clinical notes, toxicology reports and, in many editions, earlier repertories. Every step in that chain can introduce error: a symptom may have been paraphrased, a modality dropped, a remedy added on the strength of one prescriber's observation, or a name carried forward from an edition that has since been corrected.

This matters because rubric selection is where most of the clinical reasoning happens. If a rubric misstates the symptom, or if a remedy in it never produced that symptom in a proving, the case analysis is built on a false premise. Verification is not about doubting the repertory wholesale. It is about knowing which entries rest on solid primary material and which rest on a single unverified attribution.

The comparison is therefore between two different kinds of document. The repertory gives a compressed, alphabetised, remedy-weighted index. The proving gives an unedited record of what substances did to healthy volunteers, in the order it happened, with the language the provers used. Reading them side by side is the only way to see what the compiler saw, and what was left out.

Choosing Which Rubrics Are Worth Checking

Checking every rubric in a case is impractical and unnecessary. The rubrics that deserve attention are the ones carrying weight in the prescription: the strange, rare and peculiar symptoms, the modalities that narrow the field, and any rubric that is doing most of the work of distinguishing between two or three remaining remedies.

A useful filter is to ask what happens to the case if the rubric is wrong. If removing it changes nothing, it can be left alone. If removing it collapses the analysis, that rubric needs primary-source support before it is relied on. This is the same logic used in any evidence check: effort goes where the conclusion is sensitive to error.

Rubrics that are unusually specific also deserve scrutiny, because specificity is often a sign of a single observation rather than a repeated one. A rubric describing an exact time of day, a precise sensation or a highly particular circumstance may trace back to one prover or one patient. That does not make it false, but it does mean the entry should be read as a single report rather than a confirmed effect.

• Rubrics that decide the prescription rather than merely support it.
• Strange, rare and peculiar symptoms, where a single attribution can dominate.
• Highly specific rubrics naming an exact time, sensation or circumstance.
• Rubrics whose remedy list looks unusually long or unusually short for the symptom.
• Rubrics added or altered in a later edition, where editorial history is visible.

Locating the Primary Proving Record

The first practical obstacle is bibliographic. Provings are scattered across journals, monographs, provings of single remedies, and the introductory matter of older materia medica. Some are indexed; many are not. A remedy may have a nineteenth-century proving, a later reproving, and a toxicological account, each contributing different symptoms to the same rubric.

Start with what the repertory itself indicates. Many editions carry a preface or source list naming the provings and authors consulted. Where a remedy entry is drawn from a named author, that author's proving is the place to look first. Where no source is given, the symptom usually needs to be traced through the materia medica literature before it can be traced to a proving.

It helps to keep a simple record as you go: rubric, remedy, source consulted, and whether the symptom was found. Over several cases this record becomes a personal map of which parts of the repertory you have reason to trust and which remain unexamined. Nothing in the published literature substitutes for that accumulated working knowledge.

Reading a Proving for the Symptom in Question

Provings are not organised the way repertories are. Symptoms appear in chronological order within each prover's diary, mixed with the prover's own commentary, dose information, and symptoms the compiler judged irrelevant. Finding a rubric's symptom means reading for meaning rather than for a matching phrase, because the repertory wording and the prover's wording are rarely identical.

The comparison turns on several attributes. Does the proving describe the same sensation, in the same location, with the same modality and concomitant? Was the symptom produced by the proving substance or was it present before dosing? Was it reported by more than one prover, and did it recur on repetition? A symptom appearing in several provers, at a consistent time after dosing, is a different order of evidence from one passing remark.

Grade and emphasis matter here. A rubric may list a remedy in bold or in a heavier typeface to indicate frequency or intensity of the symptom in the proving. Checking the underlying record tells you whether that emphasis is justified. Occasionally a remedy is graded highly in the repertory on the strength of clinical experience rather than proving evidence, and the two should not be confused.

When the Proving and the Repertory Disagree

Disagreement is common and takes several forms. The proving may support the remedy but not the exact modality. It may support the symptom but attribute it to a different remedy that the rubric omits. It may show the symptom only as a toxicological effect at material doses rather than as a proving effect in healthy volunteers. Or the symptom may simply not be found, in which case the attribution rests on clinical observation, an earlier repertory, or an error.

Each of these calls for a different response. A modality mismatch suggests checking whether the rubric has merged two related symptoms that should be separate. An omitted remedy suggests the rubric is incomplete rather than wrong, and a case may turn on a remedy the index does not list. A toxicological-only effect raises the question of whether the symptom belongs in a proving-based repertory at all.

The hardest case is the symptom that cannot be found anywhere in the primary record. This does not automatically invalidate the rubric, because provings are incomplete and clinical experience has always informed repertories. It does mean the entry should be treated as a clinical attribution, held more lightly, and not used as the decisive rubric in a case where a better-supported alternative exists.

• Symptom found, modality different: check whether the rubric merges two symptoms.
• Symptom found, remedy absent from the rubric: treat the rubric as incomplete.
• Symptom found only at toxic doses: question whether it belongs in a proving-based index.
• Symptom found in one prover only: treat as a single report, not a confirmed effect.
• Symptom not found: treat as clinical attribution and rely on it cautiously.

Keeping a Personal Verification Record

The output of this work is not a corrected repertory. It is a working record of which rubrics you have checked, what you found, and how much weight you are prepared to give them. A short entry per rubric is enough: the symptom, the remedies that mattered, the source consulted, and a one-line verdict.

Over time the record changes how rubrics are used. Some become trusted because the proving evidence is consistent and repeated. Others become flagged, and are used only when nothing better is available. A few turn out to be more useful than expected, because the proving contains detail the repertory compressed away.

This is slow work, and it is not done once. New provings appear, editions are revised, and a remedy that was poorly proved a century ago may now have a fuller record. Rechecking an entry when a case makes it important keeps the record current without requiring a systematic audit of the whole repertory.

What Verification Can and Cannot Establish

A successful check confirms that the remedy was reported to produce the symptom in a proving, at a stated dose, in a stated number of provers. That is a statement about the source material, not about whether the remedy will help a patient with that symptom. Provings describe what substances did to healthy people; they do not predict clinical response.

Verification also cannot settle questions of rubric structure. Two compilers may reasonably divide the same clinical picture differently, and no amount of source-checking resolves a disagreement about where a symptom belongs. What it can do is make the choice explicit, so that the practitioner knows which arrangement is being followed and why.

The limits are worth stating plainly because they shape how the method is used. Cross-checking reduces the risk of prescribing on a misattributed rubric. It does not replace case-taking, and it is not a substitute for clinical judgement or for advice from a qualified practitioner. It is a bookkeeping discipline applied to the tools of prescribing.