Transition from Wooden Tablets to Paper:Exploring the Role of Mokkan as “Working Ledgers” in the Grain Loan Administration of Kaga District,Based on Materials Excavated from the Uneda–Teranaka Sites, Ishikawa Prefecture**
Introduction
This paper examines a mokkan (wooden tablet used for administrative records) bearing the date 752 (Tenpyō-Shōhō 4), excavated from the Ueda–Jichū archaeological site in Ishikawa Prefecture, and uses it as a starting point for reconsidering the practical processes through which administrative documents were produced in ancient local government.
The fundamental perspective of this paper owes much to previous scholarship, particularly the series of studies by Ryūsuke Wada. Building on this body of research, I propose the possibility of understanding this mokkan as a “working ledger” (sagyō genbo) used at the administrative site, and from this perspective seek to examine the actual processes of information handling in ancient local administration.
By “working ledger,” I mean not a completed ledger intended for final reporting and preservation, but a working record used at the operational level to record, cross-check, verify, and aggregate information such as personnel and quantities, thereby supplying the information required for the preparation of subsequent administrative accounts.
What attracts my attention in this case is not simply the information recorded on the mokkan itself. Rather, I seek to reconstruct the “logic of practice at the operational level” in ancient administration: how the mokkan was produced within a particular administrative operation, what kinds of processing it underwent, and how the information recorded on it was subsequently transferred into other ledgers and reporting documents.
A particularly important clue is a small processing mark known as a goten, which remains on the mokkan.
1. The Historical Source and the Problem
Mokkan No. 1 excavated from the Ueda–Jichū site in Ishikawa Prefecture begins with the date Tenpyō-Shōhō 4 (752). It records the names of the household head and other individuals, together with the quantities of rice corresponding to each person. The reconstructed transcription is as follows:
① 天平勝寳四年上領
② ∨戸主阿刀足人六十
③ ∨妻答尋宅女四〇
④ ∨阿刀三縄四十束
⑤ ∨妻館氣奈加女
⑥ ∨山邊足君四十
⑦ ∨□□内麻呂廿
⑧ □□悪万呂
⑨ 合稲二百□
⑩ □田秋人四十
⑫ 答尋□□女四十束
⑬ 刑部小當廿束
⑭ □姓味知麻呂十×
① Tenpyō-Shōhō 4, Jōryō
② ∨ Household head Ato Tarito 60
③ ∨ Wife Tōjin no Yashime 40
④ ∨ Ato Mitsunawa 40 bundles
⑤ ∨ Wife Tate Kinaka no Me
⑥ ∨ Yamanobe no Tarimi 40
⑦ ∨ □□ Uchimarō 20
⑧ □□ Akumaro
⑨ Total rice 200□
⑩ □□ Akihito 40
⑫ Tōjin □□ no Me 40 bundles
⑬ Osakabe no Kotō 20 bundles
⑭ □ Uji Ajimi no Maro 10×
Of particular interest are the symbols ∨, apparently goten, placed before the names in entries ②–⑦.
According to the prevailing interpretation, these marks indicate that the relevant entries had been processed or checked in the course of the accounting procedure. I agree with this interpretation.
However, simply understanding goten as a “mark indicating that an entry had been checked” does not fully exploit the information contained in this mokkan.
The more important questions are:
Why were goten marks placed before only some of the names?
Who placed them, and what exactly had been checked before each mark was added?
In what sequence did this work proceed?
And, after the checking process had been completed, where was the information transferred?
In other words, rather than treating this mokkan simply as a completed “document,” we need to read it as a record produced while a particular administrative operation was still in progress.
2. The Function of the Mokkan and Its Producer
The mokkan begins with the household head, Ato Tarito, followed by individuals who appear to have been members of his household, and records individual allocations of rice together with the total amount recorded in entry ⑨.
It therefore appears to have been more than a simple list of names. Rather, it was an administrative record used to monitor and manage the lending of rice under the suiko system at the level of individual households.
Particularly noteworthy is the fact that goten marks appear in entries ②–⑦, whereas no comparable marks appear from entry ⑧ onward.
This condition suggests that the mokkan was not a completed ledger, but rather a working record in the middle of the processing operation.
In other words, it is possible that the person responsible for the operation checked the individuals and quantities one by one and placed a goten mark beside each entry once its verification had been completed.
If this interpretation is correct, the goten was not merely a “checking mark.” It was a processing trace that recorded the progress of an administrative operation.
Equally important is the question of who required and produced such a record.
The mokkan combines several types of information—household head, household members, quantities, and a total amount. It is therefore reasonable to assume that it was produced by an official or scribal practitioner actually engaged in administering the suiko operation.
Here we should also consider the possibility that the person who held formal responsibility under the administrative system was not necessarily the same person who actually wrote the entries, checked the quantities, and placed the goten marks.
If we are to understand ancient administration not merely as a set of institutional rules but as an operation that actually functioned in practice, it is important to take such divisions of labor into account.
3. From the “Working Ledger” to Subsequent Accounts: A Model of Information Processing
Of particular relevance here is the hypothesis proposed by Ryūsuke Wada that this mokkan may have been a working record used in the preparation of paper ledgers.
This paper seeks to take that idea one step further by positioning the mokkan specifically as a “working ledger.”
During the Nara period, paper was the principal medium for official documents. At the same time, wooden tablets could have been highly effective media for the dynamic processes of collecting, checking, calculating, and correcting information at the operational level.
Accordingly, in the administration of suiko, we can envisage the following information-processing sequence:
[Collection of information at the operational level]
Names, households, quantities, and other information were recorded on wooden tablets.
↓
[Cross-checking and verification]
Individuals and quantities were checked one by one, and a goten mark was placed beside those whose entries had been verified.
↓
[Aggregation]
Quantities were totaled at the level of households, villages, districts, or other administrative units.
↓
[Reorganization into formal ledgers]
The verified information was reorganized into the formats required for subsequent paper ledgers and other administrative documents.
↓
[Submission and preservation]
The necessary information was incorporated into administrative documents submitted to higher administrative levels, such as the district and provincial governments.
↓
[End of the working ledger’s administrative function]
Once the wooden tablet had supplied the information required for the operation, it ceased to have its original administrative function.
What is important here is that we should not understand the process of “from mokkan to paper” simply as an act of copying.
The individual-level information recorded on the mokkan was not necessarily copied directly onto paper. Rather, it is more appropriate to assume that the information was aggregated and reorganized according to the purposes of the subsequent ledgers.
In this sense, the mokkan should not be understood as the “original” of a formal ledger. Rather, it can be understood as a working ledger used to generate the information required for formal administrative documentation.
4. Differentiation of Functions between Media and the “Recording of Time”
The relationship between paper and mokkan in ancient administration cannot be adequately explained simply as a case of wooden tablets serving as substitutes for paper because paper was scarce.
The two media may instead have been assigned different administrative functions.
Paper: a medium for preserving and transmitting verified information and reporting it to higher administrative authorities.
Mokkan: a dynamic working medium for collecting, organizing, checking, and calculating information at the operational level.
Once this functional differentiation is assumed, the fact that goten marks appear only in entries ②–⑦, but not beside the subsequent names, takes on considerable significance.
It does not simply mean that the document was “unfinished.”
Rather, it may indicate that the mokkan was a record that captured a particular moment in the progress of an administrative operation—a “record of time.”
In other words, the mokkan preserves one stage in a sequence such as:
record → cross-check → mark as checked → aggregate → transfer information to the next ledger
The mokkan thus preserves a stage in the middle of a process.
What remains visible on the wooden tablet may therefore be the process by which a document was being produced, something that is difficult to observe in a completed paper document.
The historical value of this mokkan consequently lies not only in what is written on it, but also in the possibility of reconstructing the temporal sequence of administrative practice from the processing traces left on its surface.
5. The Transformation of Primary Information into Aggregated Information
A closer examination of the structure of the mokkan further suggests that the process of “from mokkan to paper” involved not simply copying but also the reorganization and aggregation of information.
A single wooden tablet simultaneously records several different levels of information:
the household head;
members of the household;
quantities assigned to individual persons;
the total quantity.
These different types of information may have been used in different forms according to the purposes of subsequent administrative documents.
For example:
A. Individual suiko amounts
→ total suiko amounts by household, village, or district
→ aggregation of interest-bearing rice
→ suiko-related accounts, Shōzeichō (accounts of regular tax rice), and other fiscal records
B. Household-head, family, and personnel information
→ population data by household, village, or district
→ population records such as Keichō
In this way, information obtained from a single working record may have been reorganized into different administrative documents serving different purposes.
Here, rather than directly applying the modern concept of a “database” to ancient society, it is more appropriate to understand the structure as one in which information acquired at the individual level was reorganized into different documentary units according to the purposes of aggregation and reporting.
In other words, there may have been a hierarchy of information such as:
individual information → household information → village information → district information → provincial information
within ancient local administration.
The mokkan may have occupied the lowest level of this information-processing hierarchy: the operational site where individual-level information was collected and verified.
6. Conclusion: From the “Working Ledger” to a Model of Nara-Period Accounting
On the basis of the above considerations, I propose that the suiko mokkan excavated from the Ueda–Jichū site was not simply a memorandum or a fragmentary list of names, but was likely a working ledger used in the administration of suiko in Kaga District to record, verify, and aggregate information concerning individuals and quantities.
Of particular importance are the goten marks preserved on the mokkan.
The fact that goten marks occur only in entries ②–⑦, while they do not appear from entry ⑧ onward, may indicate that the mokkan was not a completed ledger but a record used while the administrative operation was still in progress.
If the goten marks indicate the verification of individual personnel and quantities, their placement may have followed a particular sequence.
If that sequence can be confirmed through direct examination of the original artifact, the goten would cease to be merely a symbol. It would become a trace recording the temporal sequence in which the suiko operation was actually carried out.
From this perspective, the mokkan can be positioned within the following process of administrative information processing:
Acquisition of individual information
↓
Recording on a mokkan
↓
Cross-checking of personnel and quantities
↓
Confirmation of processing through goten marks
↓
Aggregation by household, village, and district
↓
Reorganization of information into subsequent ledgers
↓
Integration into fiscal information at the district and provincial levels
Under this model, the phenomenon of “from mokkan to paper” was not simply a change of medium.
Rather, it can be understood as a process in which individual-level information recorded on a mokkan was aggregated and reorganized according to the purposes of subsequent ledgers and transformed into higher-level administrative information.
The importance of the mokkan therefore lies not merely in the fact that a wooden tablet was used as a substitute for paper.
Its greater significance lies in the possibility of reconstructing the processes through which information acquired at the individual level was transformed into the aggregated information required for administration.
This is the problem of the “generation process of administrative documents,” a dimension that has not been sufficiently visible in conventional studies of ancient documentary practices.
A completed paper ledger preserves only the final figures. Yet for those figures to come into existence, numerous names and quantities had first to be collected, checked, calculated, and aggregated.
This mokkan offers a glimpse into precisely those everyday administrative operations that existed behind the completed ledger.
Furthermore, if this perspective can be extended to other mokkan, paper documents, Shōzeichō, Gun-inchō, Keichō, and related sources, ancient local administration can be reconsidered not merely as a system of institutions, but as an information-processing system in which information was generated, processed, aggregated, and transmitted to higher administrative authorities.
From this perspective, the figures preserved in Shōzeichō and Gun-inchō need not be understood merely as “results.” They may instead be read as the final aggregated values produced from a large number of working records that once existed behind them.
We may therefore hypothesize an information flow such as:
mokkan → working records → aggregated ledgers → district finance → provincial finance → central government
This hypothesis requires further investigation.
First, the sequential nature of the goten marks must be examined.
The original mokkan should be studied in detail, comparing the form, position, and brushwork of the marks in order to determine whether they were added simultaneously or in a particular temporal sequence.
If a clear sequence can be established, it would provide an important clue for reconstructing the actual order in which the suiko operation was performed.
Second, the personal names and kinship or clan relationships require further examination.
Particular attention should be given to the household head Ato Tarito and his wife Tōjin no Yashime, including the possible relationship between the Ato clan and Ato Otari, who appears in the Shōsōin documents, as well as the problem posed by the character taku in the name of Tōjin no Yashime.
By investigating these issues, I would like to take up the challenge of approaching the social background in which this mokkan was produced and used.
Third, the physical characteristics of the mokkan itself require examination.
Precisely because I am frustrated by my inability to examine the original artifact directly, I would like to investigate in detail its material, thickness, dimensions, preparation, handwriting, condition of the ink, and use of its front and reverse surfaces.
Such examination may help determine whether the mokkan was a medium intended for preservation, or rather a working medium designed to be carried, rewritten, checked, and repeatedly consulted.
Fourth, the mokkan must be connected with the administrative structure of Kaga District.
Given that the Ueda–Jichū site appears to have had a port-related function, particularly in connection with a tsushi (port administrative office), it is necessary to examine whether suiko administration and port administration were conducted within the same administrative space or by mutually connected administrative organizations.
This may lead to what I would call a “multi-functional model of the district government office,” in which the gun-ga is understood not as a government office serving a single function, but as a space in which multiple administrative functions—including finance, transportation, and port administration—operated simultaneously and in overlapping ways.
Fifth, the information-processing model of “mokkan → paper → district finance” must be developed further.
By cross-checking the distribution of individual quantities, the sequence of goten marks, regulations governing suiko, suiko-related mokkan from other regions, and the structures of Shōzeichō, Gun-inchō, Keichō, and related records, I would like to reconstruct concretely how information acquired at the operational level was aggregated and incorporated into subsequent administrative documents.
Sixth, the ultimate goal of this research is to develop what might be called a “Nara-period accounting model”—a model of the information-processing structure underlying ancient local administration.
The term “accounting model” does not mean that modern accounting institutions are simply being projected backward onto the Nara period.
What matters is to reconstruct from the historical sources how ancient administrative practice implemented a basic information-processing structure that remains fundamental to modern accounting:
acquiring primary information
→ verifying individual information
→ aggregating it
→ transferring and reorganizing information between ledgers
→ determining the final fiscal figures
In this sense, the mokkan is a small historical source.
Yet despite its small size, a single goten mark preserved on it may have the potential to reveal the temporal sequence and working procedures of information processing in ancient administration.
When an observation beginning with a single wooden tablet illuminates the layers of information processing extending from the working ledger to paper documents, district finance, and provincial finance, we can begin to encounter something that is different from the ritsuryō system as formulated in legal texts.
We can begin to see, directly, the people who actually operated the system—people who took up their brushes, checked quantities, calculated totals, and transmitted information to the next ledger—as well as the information ecosystem that sustained ritsuryō administration.
It is precisely this world that I hope to reconstruct concretely, one trace at a time, from the surviving evidence on the mokkan.
