Every Figgipedia record follows a strict editorial doctrine: every piece of information must be traceable to a source, and the quality of that evidence is declared explicitly. The aim is to avoid the two classic pitfalls of varietal compilations — confusion through homonymy, and outright fabrication.
Why this doctrine
The fig tree suffers from a nomenclatural chaos unmatched among temperate fruit trees: more than four thousand varietal names have been recorded worldwide, for only a few hundred genetically distinct cultivars. A single variety may bear ten names depending on the region; a single name may designate cultivars that are genetically remote from one another across countries. Commercial catalogues, amateur forums and even some older scientific publications continue to circulate a substantial share of misidentifications.
Faced with this confusion, two editorial attitudes are possible. The first, dominant in French-language pomological popularisation: accept every name as it arrives, multiply records without questioning their coherence. The second, adopted here: explicitly establish the quality of each claim, refuse to publish what cannot be substantiated, and flag any residual zones of uncertainty. This is the pyramid doctrine.
The five levels of the pyramid
| Level | Criterion | Publication |
|---|---|---|
| 0 | ≥1 SSR/SNP paper AND ≥1 canonical monograph | Published |
| 1 | ≥1 SSR/SNP paper OR ≥1 canonical monograph | Published |
| 2 | Concordant secondary sources (Fig Boss, USDA-GRIN, INRAE, etc.) | Published |
| 3 | Isolated, non-canonical mention | Not published |
| 4 | Contradictory or probable homonymy (“fig trap”) | Not published |
Epistemic markers
Pyramid levels qualify a record as a whole. Within the body of a record, every factual statement is further flagged with an epistemic marker that signals the level of evidence for that specific claim. These markers are mandatory in the detailed view of every published level 0, 1 or 2 record.
- [ESTABLISHED] — multiply attested by convergent canonical sources (≥3 recent peer-reviewed papers or a reference monograph)
- [PROBABLE] — plausible, supported by 1 to 2 solid sources or by reasoned analogy with a closely related species
- [UNCERTAIN] — needs confirmation, single or weak source, cautious extrapolation
- [CONTESTED] — contradictory sources without arbitration, active scientific debate
This open transparency on levels of evidence sets Figgipedia apart from other varietal compilations of the fig tree. Readers know on every line whether they are reading a solid consensus or a claim still awaiting verification — and can judge for themselves the weight to give to each assertion.
The “fig traps” — automatic level 4
Some situations automatically trigger the maximum doubt level:
- Purely descriptive names (“Black Fig”, “Figue Noire”, “Green Fig”)
- The same name used in different regions without genetic confirmation
- Divergent functional type (Common vs Smyrna, for instance)
- Name declared at acquisition without SSR verification
A concrete example — the Goutte d'Or case
Goutte d'Or is one of the most cited French heritage varieties in garden nurseries. Yet analysis reveals that at least four genetically distinct lineages circulate under this name depending on the region (Aquitaine, Vaucluse, Rhône Valley, Charentes), each with different morphological and organoleptic profiles.
A naïve compilation would produce a single “Goutte d'Or” record accumulating contradictory information. The Figgipedia doctrine requires, over time, as many records as there are documented lineages — each classified according to its own SSR/SNP evidence, and each explicitly flagging [UNCERTAIN] the attribution of the commercial name “Goutte d'Or” to that precise lineage. This is no small matter: it is the difference between a catalogue that says “here is what is sold under this name” and an encyclopaedia that says “here is what science can confirm”.
Evolution over time
The pyramid is not fixed: a level 2 record (concordant secondary sources) can be promoted to level 1 as soon as an SSR/SNP study is published for the variety. Conversely, a record can be demoted if a recent publication invalidates an earlier attribution. The modification history at the foot of each record traces these evolutions — a transparency rarely found in compilatory pomology.