TheRON/otivm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add historical-reality-parameters.md

Browse Source 
schema instrument for enslaved labour, legal discrimination, commercial sex, and public violence
This commit is contained in:
TheRON
2026-04-29 05:39:39 -04:00
parent cfac01216c
commit b5b522e1e1
1 changed files with 629 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

629
docs/architecture/historical-reality-parameters.md Normal file
View File

@@ -0,0 +1,629 @@
# Historical Reality Parameters
### TheRON — OTIVM / CIVICVS Parameter Schema Instrument
### Status: Internal only — not player-facing content
### Date: 2026-04-28
---
## 0. Purpose and Scope
This document is an internal schema instrument. It is not a scenario.
It is not player-facing content. It will not be shown to participants.
Its purpose is to map the parameter domains required for an economically
honest simulation of Roman commercial life in approximately 14 BCE. Four
domains are covered:
1. Enslaved labour
2. Legal and status discrimination
3. Commercial sex
4. Public violence and the arena
These domains are not peripheral to Roman economic life. They are structural.
A simulation that excludes them cannot model Roman economics accurately.
A simulation that includes them as gratuitous spectacle has failed its purpose.
**The principle governing this document:**
These domains are modelled as parameters and economic forces. The simulation
does not editorialise. It models. The participant encounters these as the
MERCATOR encounters them — as the texture of the world they operate in,
not as moral choices presented for approval. The facts no longer affect
living beings. They belong to a historical period whose actors have all
ceased to exist. The purpose of modelling them is contribution to accurate
historical understanding, not sensation.
**Method:** for each domain, this document identifies:
- The structural role of the domain in the Roman economy
- Existing parameters from `docs/architecture/parameter-registry.md` that
are already affected by this domain, with the nature of the effect stated
- New parameters required that are not yet in the registry
- Sources
---
## 1. Enslaved Labour
### 1.1 Structural role
Slavery was not an anomaly in the Roman economy. It was its operating
system. Estimates suggest enslaved persons constituted 15–30% of the
Italian population in the late Republic and early Empire — potentially
2–3 million people in Italy alone. In the commercial harbour economy of
Ostia, enslaved labour was present at every operational level: BAIВLVS
(porters), warehouse workers, ship crew members, accounting staff,
household service, and skilled artisans.
The MERCATOR operated inside this system whether or not he personally
owned enslaved persons. He hired them, contracted their labour through
their owners, competed against their output, and used infrastructure
they maintained. His FACTOR might be enslaved or freedman. His access
to certain services depended on this system.
The Roman economy did not develop water-powered or steam-powered
industrial production at scale despite possessing the engineering
knowledge to do so. The reason is documented: human bodies were cheaper.
This is not an inference — it is the economic consequence of a labour
cost structure where the marginal cost of additional enslaved labour
was lower than the capital cost of mechanical substitution. This
consequence must be present in the parameter model.
**Primary sources:**
- Cato the Elder, *De Agricultura* — management of enslaved agricultural
workers, cost structures, maintenance calculations
- Columella, *De Re Rustica* — detailed labour cost accounting
- Digest of Justinian, Book 21 — ACTIO EMPTI and sale of enslaved persons,
warranty obligations, disclosure requirements
- Varro, *Rerum Rusticarum* — classification of labour as *instrumentum
vocale* (speaking tools), *semivocale* (semi-speaking: animals),
*mutum* (mute: inanimate)
**Secondary sources:**
- Keith Hopkins, *Conquerors and Slaves* (1978)
- Moses Finley, *The Ancient Economy* (1973)
- Walter Scheidel, *The Cambridge Economic History of the Greco-Roman World*,
Chapter 5 (2007)
### 1.2 Existing parameters affected
**`liquiditas`**
A MERCATOR who owns enslaved workers holds capital in a non-liquid form.
Their sale value is an asset; their maintenance is a recurring cost.
The ratio of enslaved-to-hired labour in a commercial operation directly
affects the owner's liquidity profile. The existing `liquiditas` parameter
must accommodate a distinction between liquid capital and capital held in
human assets.
**`labour_cost` (stub — see §1.3)**
Currently absent from the registry. Hired free labour (MERCENNARIUS)
and contracted enslaved labour have different cost structures, different
legal exposure for the contracting party, and different reliability
profiles. These cannot be collapsed into a single cost parameter.
**`ius_accessus`**
An enslaved person has no IVS_ACCESSVS in Roman law. They cannot
enter contracts, appear as witnesses, or initiate legal proceedings
in their own name. A MERCATOR conducting NEGOTIA through an enslaved
FACTOR (institor servilis) has a specific legal exposure profile:
the owner is liable for the FACTOR's commercial acts up to the value
of the PECULIUM (the allowance granted to the enslaved person for
commercial use). This is the *actio institoria* and *actio tributoria*
framework. The `ius_accessus` differential between a CIVIS, a LIBERTUS,
and a SERVUS is not a single ordinal scale — it is a legally structured
set of distinct capabilities and incapacities.
**`auctoritas`**
A MERCATOR who treats enslaved persons visibly well or badly affects
their AVCTORITAS differently depending on the social context. Excessive
cruelty was considered poor form even in a slave-owning society —
not on humanitarian grounds, but because it signalled poor management
and social instability. The AVCTORITAS system must accommodate the
social signalling dimension of how an actor manages labour.
**`officia_burden`**
Ownership of enslaved persons creates legal obligations (maintenance,
the *actio de peculio*, liability for their commercial acts) that
contribute to OFFICIA_BVRDEN. This is not modelled in the current
registry entry, which frames OFFICIA_BVRDEN primarily in terms of
social obligations.
### 1.3 New parameters required
**`labour_source`**
```
token: labour_source
scope: actor
layer: roman
maturity: research_needed
```
The composition of an actor's labour force: proportion enslaved vs
free hired vs freedman contracted. Affects `liquiditas` profile,
`ius_accessus` exposure, legal liability, and operational flexibility.
An actor with primarily enslaved labour has lower variable costs but
higher capital locked in assets and higher legal exposure for their
acts. An actor with primarily hired free labour has higher variable
costs, lower asset lock, and cleaner legal separation.
**`labour_cost`**
```
token: labour_cost
scope: scenario
layer: roman
maturity: research_needed
```
The cost per unit of labour for a specific ITER or operational task,
disaggregated by labour type. BAIВLVS day rate for free hired labour
is documented in Diocletian's Edict (301 CE, later than our period but
provides relative structure). Earlier estimates require interpolation
from Cato and Columella. Research needed: Ostia-specific rates,
1st c. BCE.
**`peculium_value`**
```
token: peculium_value
scope: actor
layer: roman
maturity: research_needed
```
The commercial allowance granted to an enslaved FACTOR for use in
NEGOTIA. Sets the ceiling of the owner's liability under *actio
tributoria*. Also the de facto working capital of an enslaved
commercial agent. A FACTOR with a large PECULIUM is effectively
conducting independent commercial operations on behalf of their owner.
This is one of the mechanisms by which enslaved persons could
accumulate capital toward self-purchase (MANUMISSIO).
**`manumission_probability`**
```
token: manumission_probability
scope: actor
layer: roman
maturity: research_needed
```
The probability that a skilled enslaved commercial agent will be
formally freed (MANUMISSIO) within a defined time horizon, converting
to LIBERTUS status. Relevant because LIBERTUS actors have a different
parameter profile from both CIVIS and SERVUS actors. The transition
from SERVUS to LIBERTUS is a background drift event of the highest
magnitude — it changes IVS_ACCESSVS, AVCTORITAS floor, and social
network access simultaneously. This transition is the origin story of
the BACKGROUND-0002 (Freedman Trader) cast profile.
---
## 2. Legal and Status Discrimination
### 2.1 Structural role
Roman society was organised around legally encoded status hierarchies
that directly governed commercial capability. These were not informal
prejudices — they were written into law and enforced by courts. The
relevant distinctions for the MERCATOR's world:
**CIVIS ROMANVS** — full Roman citizen. Full legal capability: contract,
witness, property ownership, legal action. The baseline.
**LATINVS** — Latin status. Commercial rights but restricted political
and full legal rights. Many LIBERTI were LATINI IVNIANI — freed but
without full citizenship.
**PEREGRINUS** — foreign free person. Commercial activity permitted
under *ius gentium* but restricted Roman law access. Significant in
Ostia, a port city with large foreign populations.
**LIBERTUS / LIBERTA** — freedman/woman. Citizen status in most cases
(if freed formally by a CIVIS) but socially marked by servile origin.
Could not hold most public offices. Subject to ongoing *operae* (labour
obligations) to former owner. In practice, freedmen dominated Roman
commercial and craft activity.
**SERVUS** — enslaved person. No legal personhood. No contract, no
witness, no property (technically — PECULIUM was a practical workaround).
**Women** — regardless of status, Roman women had restricted commercial
legal capability without a male guardian (TVTOR) in most circumstances.
Exceptions existed; they were exceptions.
**Non-Roman ethnic and religious communities** — Jews, Egyptians, and
other identifiable groups faced specific restrictions, social hostility,
and periodically legal exclusions that affected commercial activity.
This is not modern racism but it had comparable commercial effects:
restricted access to certain markets, inability to use certain legal
instruments, exclusion from some COLLEGIA.
**Primary sources:**
- Gaius, *Institutiones* — systematic treatment of legal status categories
- Digest of Justinian, Books 1, 4, 40 — citizenship, manumission, legal capacity
- Cicero, *Pro Balbo* — citizenship as commercial prerequisite
**Secondary sources:**
- Jane Gardner, *Being a Roman Citizen* (1993)
- A.N. Sherwin-White, *The Roman Citizenship* (1973)
- John Bodel, *Epigraphic Evidence* — freedman commercial activity
### 2.2 Existing parameters affected
**`ius_accessus`**
The existing registry entry describes this as an ordinal scale (low /
medium / high). This is insufficient. IVS_ACCESSVS is not a continuous
variable — it is a structured set of legal capabilities that differ
categorically between SERVUS, LATINVS, PEREGRINUS, LIBERTUS, and CIVIS.
The schema must accommodate legal status as a discrete category, not
an ordinal score. The ordinal representation is a simplification that
will produce wrong results in legal dispute scenarios.
**`auctoritas`**
The existing registry notes AVCTORITAS as partially observable and
socially constructed. The legal status layer adds a floor and ceiling
to AVCTORITAS that is structurally imposed, not just socially earned.
A LIBERTUS cannot exceed a certain AVCTORITAS threshold regardless of
commercial success, because certain social expressions of AVCTORITAS
(holding office, certain COLLEGIA membership) are legally closed to him.
A CIVIS of low commercial achievement still has a higher AVCTORITAS
floor than a successful LIBERTUS in formal legal contexts.
**`information_quality`**
Access to commercial information in Rome was heavily mediated by
social networks that were themselves status-stratified. A PEREGRINUS
in Ostia had access to information networks within his ethnic community
but reduced access to Roman citizen networks. A LIBERTUS had access
to his former owner's network (CLIENTELA) but was excluded from others.
The `information_quality` parameter must accommodate network access
constraints derived from legal and social status.
**`negotiatio`**
Negotiation capability was not purely a personal skill — it was
mediated by legal standing. A SERVUS negotiating on behalf of an
owner could achieve certain outcomes but was legally constrained in
others. A PEREGRINUS negotiating with Roman citizens was operating
under *ius gentium*, not Roman civil law, with different remedies
available if the counterparty defaulted.
### 2.3 New parameters required
**`legal_status`**
```
token: legal_status
scope: actor
layer: roman
maturity: canonical
```
The actor's formal Roman legal status. Discrete categories, not ordinal:
`civis_romanus`, `latinvs`, `peregrinus`, `libertus`, `liberta`, `servus`.
This is the foundational parameter that determines the structure of
`ius_accessus` for each actor. It does not drift — it changes through
specific legal events (manumission, citizenship grant) which are
recorded as events in the time-series.
**`ethnic_community`**
```
token: ethnic_community
scope: actor
layer: roman
maturity: provisional
```
The actor's cultural and ethnic community affiliation, where distinct
from Roman citizen status. Affects `information_quality` within community
networks, access to community-specific commercial infrastructure (e.g.
Jewish merchant networks, Egyptian grain traders, Syrian traders in
Ostia), and exposure to community-specific legal restrictions and social
hostility. Not a racial category — a social and legal one.
**`tutor_required`**
```
token: tutor_required
scope: actor
layer: roman
maturity: provisional
```
Boolean. Whether the actor legally requires a male guardian (TVTOR)
to execute certain commercial transactions. Applies to women actors.
Affects `ius_accessus` for specific transaction types. A woman MERCATRIX
(female merchant — attested in inscriptions) conducting NEGOTIA without
a TVTOR faces legal exposure on certain contract types.
---
## 3. Commercial Sex
### 3.1 Structural role
Commercial sex in Rome was a licensed, taxed, legally categorised
industry. This is not a peripheral fact — it is documented in
municipal records, legal texts, and archaeological evidence across
the Roman world including Ostia.
Key structural elements relevant to the simulation:
**MERETRIX** — a woman registered as a prostitute with the AEDILE
(municipal official). Registration was required by law (Lex Iulia
de adulteriis, 18 BCE — near our simulation period). It conferred
a specific legal status: the MERETRIX was exempt from adultery law
(applying only to respectable women) but permanently infamis —
legally disgraced, stripped of certain legal protections.
**LENO / LENA** — the brothel-keeper (male / female). A commercial
operator. Subject to the INFAMIA legal sanction, which removed certain
legal protections and rights. Taxed: the VECTIGAL MERETRICVM was a
municipal revenue source.
**FORNIX / LUPANAR** — the physical location. In Ostia, archaeologically
attested in harbour and market districts — precisely the areas the
MERCATOR operates in.
**Commercial relevance to the MERCATOR:**
- The LUPANAR was part of the harbour economy. Its proximity to docks,
warehouses, and taverns is not coincidental — it served transient
labour populations: sailors, porters, travelling merchants.
- Expenditure in this sector is a legitimate economic flow that affects
the actor's `liquiditas` and OFFICIA_BVRDEN profile.
- The LENO as commercial actor had specific legal constraints that
affected contract enforceability.
- Association with this sector affected AVCTORITAS and FAMA differently
depending on the actor's status and the visibility of the association.
**Primary sources:**
- Digest of Justinian, Book 3.2 — INFAMIA, legal consequences of
MERETRICIA profession
- CIL (Corpus Inscriptionum Latinarum) — Pompeii price lists (later
than our period but structurally informative)
- Plautus comedies — commercial and social references (earlier period,
informative for social norms)
**Secondary sources:**
- Thomas McGinn, *Prostitution, Sexuality, and the Law in Ancient Rome* (1998)
- Sarah Levin-Richardson, *The Brothel of Pompeii* (2019)
- Rebecca Flemming, *Medicine and the Making of Roman Women* (2000)
### 3.2 Existing parameters affected
**`liquiditas`**
Expenditure in this sector is a recurring cost parameter for actors
whose social patterns include harbour district activity. It is
economically indistinguishable from expenditure on food, wine, or
lodging in the simulation's accounting model — it is a cost that
reduces available capital. It must not be treated as categorically
different from other consumption costs in the schema.
**`auctoritas`**
Visible association with the sector affected AVCTORITAS differently
by social class. A high-status actor (Lentulus) visible in a LUPANAR
faces AVCTORITAS and FAMA damage. A low-status actor (Felix, Varro)
faces no particular consequence — it was unremarkable for their
social level. This is a `perceived_vs_true` interaction: the social
cost is mediated by who observes, not by the act itself.
**`fama`**
The FAMA parameter must accommodate sector-specific social visibility.
Certain locations and associations produce FAMA effects only if observed
by status-relevant witnesses. The BALNEA rumour network
(`rumor_velocity`) carries some of this information; the schema must
model that rumour content has differential impact by observer status.
**`officia_burden`**
Not applicable for the MERCATOR as consumer. Applicable if modelling
the LENO as a commercial actor type — ownership of a LUPANAR creates
specific legal and operational obligations.
### 3.3 New parameters required
**`infamia_flag`**
```
token: infamia_flag
scope: actor
layer: roman
maturity: canonical
```
Boolean. Whether the actor carries the legal status of INFAMIA —
legal disgrace that removes certain protections and rights. Applies
to: LENO, LENA, MERETRIX, gladiators, actors, and others defined by
Roman law. Affects `ius_accessus` (certain legal actions barred),
`auctoritas` (formal ceiling reduced), and `clientela` (certain
patron-client relationships unavailable). Not the same as low FAMA —
INFAMIA is a legal status, FAMA is a social perception.
**`sector_visibility`**
```
token: sector_visibility
scope: relation
layer: roman
maturity: provisional
```
The probability that an actor's commercial or social activity in a
specific sector is observed by status-relevant witnesses and enters
the rumour network. Modulates `fama` effects for activities that have
differential social consequences by observer. High `sector_visibility`
in the harbour district late at night produces different FAMA effects
than the same activity unseen.
---
## 4. Public Violence and the Arena
### 4.1 Structural role
Public violence in Rome was not an anomaly or a failure of civilization.
It was a deliberately maintained social institution with specific
economic, political, and social functions. The relevant forms for the
MERCATOR's world:
**LVDI** — public games, including gladiatorial combat (MVNERA),
animal hunts (VENATIONES), and chariot racing (LUDI CIRCENSES).
Sponsored by magistrates, wealthy patrons, and the Emperor. The sponsor
(EDITOR MVNERUM) gained AVCTORITAS and CLIENTELA in proportion to the
scale and quality of the event. This was not entertainment in the modern
sense — it was a mechanism for redistributing social capital from
wealthy to popular, cementing political alliances, and demonstrating
the sponsor's power and generosity.
**Gladiatorial schools (LVDI GLADIATORUM)** — commercial operations.
A LANISTA (gladiatorial trainer/manager) purchased, trained, and hired
out gladiators. This was a profitable but INFAMIA-carrying business.
Gladiators themselves were often enslaved persons or condemned criminals,
but also included free volunteers (AVCTORAMENTVM) drawn by pay, status
within the arena world, and — occasionally — the release from debt.
**Capital punishment** — public execution (SVPPLICIUM) was a regular
civic event. Criminals, slaves, and enemies of the state were executed
publicly in ways that served as both deterrent and spectacle. This is
not tangential to the commercial world — condemned criminals included
defaulted debtors in some circumstances.
**Commercial relevance to the MERCATOR:**
- Attending, sponsoring, or being seen at LVDI was an AVCTORITAS and
CLIENTELA event, not merely entertainment.
- The LANISTA as commercial actor operated within specific legal
constraints (INFAMIA) while conducting profitable trade in human
combat capacity.
- The economic demand generated by LVDI — for animals, equipment,
food, temporary labour, lodging — was a commercial opportunity the
MERCATOR could service.
- The EDITOR MVNERUM role was a form of `liquiditas` expenditure that
produced `auctoritas` returns — one of the clearest Roman examples
of converting money into social capital.
**Primary sources:**
- Suetonius, *Lives of the Twelve Caesars* — sponsorship economics,
scale of games
- Cicero, *Pro Sestio*, *De Officiis* — AVCTORITAS logic of public
sponsorship
- CIL — inscriptions recording EDITOR MVNERUM, costs, and social returns
- Digest of Justinian, Book 3.2 — INFAMIA of LANISTA
**Secondary sources:**
- Keith Hopkins and Mary Beard, *The Colosseum* (2005)
- Fik Meijer, *The Gladiators* (2003)
- Donald Kyle, *Spectacles of Death in Ancient Rome* (1998)
### 4.2 Existing parameters affected
**`auctoritas`**
The EDITOR MVNERUM mechanism is the clearest Roman example of direct
`liquiditas` → `auctoritas` conversion. The existing registry entry
does not describe this pathway explicitly. It must be added: spending
on public sponsorship (LVDI, public feasts, building dedications) is
a direct investment in AVCTORITAS that bypasses the slow accumulation
through repeated commercial reliability. It is also irreversible —
a patron who sponsors games cannot unsponsor them, and the AVCTORITAS
gained decays if not maintained by further sponsorship.
**`clientela`**
Public games generated CLIENTELA directly. Attendees who received
free admission or gifts (MISSILIA — tokens thrown to the crowd)
entered a diffuse client relationship with the sponsor. This is a
mass CLIENTELA acquisition mechanism — distinct from the individual
cultivation described in the existing registry entry. The schema must
accommodate both forms.
**`officia_burden`**
The EDITOR MVNERUM role created specific obligations: to the audience
(expectation of future games), to the performers contracted, to the
animals suppliers, to the venue. Once entered, the role generated
ongoing OFFICIA_BVRDEN whether or not the actor wished to continue.
**`liquiditas`**
Sponsoring LVDI was a major capital expenditure. The scale ranged from
modest municipal games to the multi-day spectacles of the Imperial
period. For a working MERCATOR, even modest sponsorship represented
a significant `liquiditas` commitment with an uncertain `auctoritas`
return — because the social capital gained depended on attendance,
weather, the quality of the performance, and the political climate.
### 4.3 New parameters required
**`sponsorship_investment`**
```
token: sponsorship_investment
scope: actor
layer: roman
maturity: provisional
```
Capital committed by an actor to public sponsorship (LVDI, public
feasts, building dedications). Produces `auctoritas` return at a
rate modulated by event scale, attendance, and political climate.
The return is delayed (AVCTORITAS accrues after the event, not during
the expenditure) and uncertain (bad weather, poor performance, or
political interference reduces return). Irreversible once committed.
**`arena_demand_index`**
```
token: arena_demand_index
scope: city
layer: roman
maturity: provisional
```
The current commercial demand generated by scheduled or recent LVDI
in a city — for animals, equipment, food, lodging, and temporary
labour. A city in the week before major games has different commercial
conditions from a city in a quiet period. Affects `food_price_index`,
`porter_availability`, `storage_fee_index`, and `dock_congestion`.
**`lanista_flag`**
```
token: lanista_flag
scope: actor
layer: roman
maturity: provisional
```
Boolean. Whether the actor is operating as or in partnership with a
LANISTA (gladiatorial school operator). Applies INFAMIA consequences
(see `infamia_flag`) while enabling access to specific commercial
networks (arena supply chains, condemned labour pools). A commercial
actor who avoids this sector loses certain opportunities; one who
enters it gains opportunities at AVCTORITAS cost.
---
## 5. Cross-Domain Parameter Summary
New parameters introduced by this document, for addition to
`docs/architecture/parameter-registry.md`:
| Token | Scope | Layer | Domain | Maturity |
|---|---|---|---|---|
| `labour_source` | actor | roman | enslaved labour | research_needed |
| `labour_cost` | scenario | roman | enslaved labour | research_needed |
| `peculium_value` | actor | roman | enslaved labour | research_needed |
| `manumission_probability` | actor | roman | enslaved labour | research_needed |
| `legal_status` | actor | roman | legal discrimination | canonical |
| `ethnic_community` | actor | roman | legal discrimination | provisional |
| `tutor_required` | actor | roman | legal discrimination | provisional |
| `infamia_flag` | actor | roman | commercial sex / arena | canonical |
| `sector_visibility` | relation | roman | commercial sex | provisional |
| `sponsorship_investment` | actor | roman | arena | provisional |
| `arena_demand_index` | city | roman | arena | provisional |
| `lanista_flag` | actor | roman | arena | provisional |
**Existing parameters requiring schema revision:**
| Token | Required revision |
|---|---|
| `ius_accessus` | Must become a structured legal capability set, not an ordinal scale. Keyed to `legal_status`. |
| `auctoritas` | Must accommodate: status-imposed floor and ceiling; direct `liquiditas`→`auctoritas` conversion via sponsorship; INFAMIA ceiling constraint. |
| `officia_burden` | Must include: liability for enslaved persons' commercial acts; sponsorship obligations once entered. |
| `information_quality` | Must accommodate: network access constraints derived from `legal_status` and `ethnic_community`. |
| `fama` | Must accommodate: `sector_visibility` modulation — FAMA effects are observer-dependent, not universal. |
| `liquiditas` | Must distinguish: liquid capital vs capital held in human assets (enslaved persons). |
---
## 6. Schema Discipline
These parameters must not be encoded differently from any other parameter
in the schema. They are historical facts about a specific period and
place, not editorial content. The same precision, confidence tagging,
and source citation standards apply here as to cargo weights and customs
duties.
Uncertainty is a first-class record. Where source data is thin —
particularly for labour cost rates and PECULIUM values specific to
Ostia in 14 BCE — the parameter is marked `maturity: research_needed`
and the confidence tag reflects the gap. A gap honestly documented is
more useful than a false precision.
---
*Historical Reality Parameters — internal instrument, 2026-04-28*
*Not player-facing. Schema use only.*
*The simulation models. It does not editorialise.*
*TheRON — single contributor. AI assistants implement, document, flag — do not direct.*