A Few Notes on Wood Claims in China

Spent the past few days reading collections of “popular customs” (民事习惯) that were compiled during the push to compile a civil law code (民法) in late Qing and Republican China, circa 1903–7 and 1917–30. This was a point when dozens of longstanding practices were written down, giving a rare widow onto the broader world of claims (or rights) and responsibilities. It is a bit dangerous to overgeneralize or to read these too far back into history, although I have found numerous other sources (principally collections of deeds and contracts) that tend to confirm the basic picture. In any case this give us a window onto the sheer variety of ways in which claims to property were divvied up — largely without direct support or interference from state agents.

Before getting into observations on the ways in which claims were divvied up, it is worth a few clarifications of language.

Terms for contracts and the people who wrote them

First, it is worth noting a plethora of different terms for landlord (zhu 主，yuanzhu 原主，maizhu 賣主, etc.), tenant (chengdian ren 承典人／承佃人，dianke 佃客，maizhu 買主), the act of leasing (maimai 買賣，diandang 典當，dian 佃，zu 租), and even the name of the deed/contract (qi 契，yue約, piao 票，zi 字， and various compounds and less common terms).

There are some semantic differences between these, especially between the notion of “sale” (mai 賣) and “rental” (zu 租, dian 佃), with “conditional sale” or “pawn” (dian 典, dianmai 典賣) theoretically splitting the difference. In practice, the “original owner” (yuanzhu 原主) almost always retained some claims to land, regardless of whether the word on the deed specified “sale” or “rental,” and the attempt to parse between “final sales” (juemai 絕賣， dumai 杜賣) and “conditional sales” (dianmai) in the deed title has little bearing on the actual division of rights/claim.

Nor does the name of the contract have much bearing on its content — with the exception of hetong 合同, which almost always means a contract where rights/claims are clarified rather than transferred; and fenshu 分書, which almost always means a will — the other terms for deed/contract were essentially interchangeable , evan as local practice tended to specify different terms for different types of deeds.

There were also a number of terms for middlemen — both those arranging and overseeing deals (zhongren 中人，yaren 押人, etc.) and those writing the contracts (代筆人 etc). Again, the name appears to have mattered less than the description of their role.

As the compilers of the reports on “people’s customs” repeatedly note, instead of focusing on the titles for deeds, we must take into account the wording of the body of the documents, and the unspoken customs that lay behind them.

In the name of simplicity, I will call all those with older or more fundamental claims “landlords,” those with newer or more superficial claims “tenants,” and the transfer of rights/claims a “lease” or “rental.” I will call all documents “contracts.”

On the other hand, the compilers of the reports are themselves generally consistent in their usage: zhu or yuanzhu for landlord, chengdianren for tenant, qi for deed, etc. As a further example, there do seem to be important differences between “deeds” (qi 契), “receipts” (piao 票) “[private] account-books” (bu 簿), and “[semi-official] registers” (ce 册), that I will get into later.

On “rights” versus “claims”

Another area where we must be careful is to differentiate the ways in which the compilers of the reports want property rights/claims to work, and the way they work in practice at that point in time.

The compilers tend to talk in terms of “rights” (quan 權), almost certainly under the influence of Western and/or Japanese discourses; indeed the sections of these reports I use here are titled “rights to goods” (wuquan 物權) often with specific reference to “immovable property” (budong chan 不動產, aka “real estate”).

<tangent> In my understanding, the premodern use of quan 權 is largely in relation to the state, and specifically to what are commonly called “monopolies.” In other words, the historical use of the term quan is largely related to exclusive claims — as in the state’s exclusive claims to goods like salt, iron and tea.

<subtangent> These exclusive claims were manifest in many different ways, including what we would strictly call monopolies, but also what should more properly be termed monopsonies or licensing systems. </subtangent>

The point being that these types of exclusive claims were largely confined to the state, at least until the very late Qing, and did not bear immediate consequence for the ways in which these claims were apportioned. Until at least the late 19th century, “rights” doesn’t work super well. I assume that around the late Qing jinshi 經世 movement, the term quan was reinterpreted, probably under influence from the West by way of Japan.</tangent>

In the deeds themselves, however, the language used is almost always “possession of” (suoyou 所有, yongyou 擁有) or “benefit from” (xiangyou 享有) or “permitted to” (ke 可) on the one hand, and “non-interference” (無礙) or “not permitted to” (buke 不可) on the other. For future exchanges (for example, of rent) both parties tend to note their “willingness” (qingyuan 情願) to do these things. More to the point, the “possession” of a good did not necessarily imply unconditional ownership or use — conditions that tended to be hashed out in contracts only when they became problems. Finally, contracts almost always specified a basis of evidence (ping 憑) — often the contract itself, but sometimes taking other forms.

In my understanding, contracts were essentially formalizations of an agreement that one party (i.e. the “landlord”) would transfer possession of a good (land) to another party (i.e. the “tenant”), sometimes with specific terms for what was permitted to each party, and with expressions of willingness by one or both parties to any future obligations; in cases where either party wanted to make a new claim, the contract also presented a basis of evidence to help resolve it. There is little to no discussion of exclusive rights. This may be splitting hairs, but I will try to reserve my discussion to these terms, rather than the discourse of rights…

Types of terms in woodland contracts

We can find six fundamental terms:

1. The good transferred to the possession of the tenant, and sometimes a specification of the good retained by the landlord.
2. The form of payment from the tenant to the landlord
3. The term of the lease
4. The basis of evidence to resolve claims
5. Certain actions permitted to each party
6. Who is responsible for outside obligations (such as taxes) on the property

These are in roughly declining order of importance and frequency. Almost all cases of tenancy report 1 and 2, many report 3 and 4, and some report 5 and 6. But as noted above, the last 4 items tend to only be reported if there are observed to be problems resolving them.

Basic terms of tenancy

At the moment, I’m interested in woodland resources, so the main goods involved are the land where wood grows, which in the Chinese context is called a shan 山 or shanchang山場 — literally a “mountain” or “mountain plot,” and any wood on that land, generally termed zhumu 竹木 (bamboo and timber) or shu 樹 (trees), and sometimes specifying specific types of trees.

There are two fundamental ways that these goods were divided between the two parties in a lease agreement.

First, the tenant pays an annual rent, and effectively takes possession of the mountain plot any any wood on it. The original documents tend to call this “mountain skin” 山皮 and the secondaries call it “surface rights.” The landlord retains some underlying claim, often called “mountain bones” 山骨 in the primaries and “subsurface rights” in the secondaries. This is, in many ways, the most straightforward form of tenancy.

I have found one context (Le’an, Jiangxi 江西樂安) where the “original owner” could sell the “bones” and retain the “skin” — in other words, become a tenant on his own land. This was essentially a form of mortgage, where a second party paid a lump sum in exchange for annual rent payments, but the first party continued to occupy the land and retained the most fundamental claim. But this appears to have been very unusual.

Second, the tenant and the landlord agree to split the profits from the sale of wood (or other products). This is essentially a term similar to “sharecropping,” except that the time it takes to grow lumber forces complications.

Both of these forms of tenancy were very, very common almost everywhere. There is nothing in the sources to suggest to me that sharecropping was more common at the frontiers and rents in core commercial regions, or vice-versa. Indeed, there are many counties specifically noted for both forms of tenancy.

On to the complications:

Duration

Both fixed-term and unlimited term tenancy were very common, under both annual rents and sharecropping arrangements. Fixed terms varied from 3–5 years all the way to 20 years or more. So-called “permanent tenancy” was also very common.

There is a huge variation here, but two patterns emerge:

1. Timberland was very often under very long-term or permanent tenancy. Annual crops — including tea, tea oil, tung, grains and tubers — were more often under short-term tenancy.
2. Permanent tenancy was very common (but not universal) at frontiers or in cases where the tenant was central to the process of “opening” the land （kaiken 開墾, kenzhong 墾種).

In addition, there was some variation on what happened at the end of a fix-term tenancy. The most common pattern was that all the wood (or other goods) on the surface of the mountain plot belonged to the tenant. There were three variant ways of dealing with this at the beginning and end of the contract:

1. The tenant payed the landlord an additional amount at the beginning of term for the value of any trees already on the land.
2. The landlord payed the tenant back for the value of any trees left on the land at the end of term.
3. The tenant clear-cut the plot at the end of the term.

Boundaries

“Mountain” plots often followed the standard system of recording four boundaries (sizhi 四至) — i.e. North-East-South-West, or upper-lower-left-right — and a location (zuoluo 坐落) within the system of villages and districts etc. This appeared to account for the majority of contexts. But more often than agricultural land/lowland, mountains had non-standard systems for reporting boundaries and location.

These included:

1. Only the name of the mountain indicated, without boundaries or a location in the village system.
2. Only a local location name was given (i.e. not a place in the village system of place-name hierarchies).
3. Boundaries relying on neighbors to confirm.
4. Old trees used as boundary markers
5. Graves used as boundary markers

Origins of claims

“Original” claims to land were strikingly persistent, even after supposedly “final” sales (dumai, duanmai). But for the most part, the source of the claims by the original owner (yuanzhu 原主) are not noted. What do we know about the origins of the claims that are noted?

In regions that were formerly peripheral or depopulated by war, the original claims often lay with the persons making the first claim to unclaimed land (無主之土 etc.). Sometimes there were specific institutional histories. In Hunan, depopulated in the Yuan-Ming, and to a lesser degree the Ming-Qing transitions, first settler claims mattered a great deal.