IntroductionFootnote 1
To the best sister in the world,
Magdalene
The discovery of Linear B documents at the Late Bronze Age palace of Pylos in mainland Greece led the first excavators to propose the existence of an archive complex (AC) within this palatial centre.Footnote 2 At Pylos, most of the tablets have well-documented find-spots, stratified within the palace's final destruction layer, dated to the LH IIIB–C transition (around 1200 BCE). About 80% of the 1300 tablets at the site were recovered from Rooms 7 and 8, where researchers argue for the in-situ systematic filing and storage of Linear B documents written by different scribes. To date, the use of the term ‘archive’ for the interpretation of the tablets’ find-spots in these areas remains a general principle in the international bibliography.Footnote 3 There are also 200 Linear B records from other parts of the palace, out of which the most numerous collection was in the Northeastern Building (NEB). Many of these tablets, originally written elsewhere within the palatial complex, were probably transferred later to the AC for further modifications or storage.Footnote 4 Nevertheless, the deposits of inscribed documents in Rooms 7 and 8 may not necessarily belong to archives. It is equally possible that the Linear B records found here may constitute a different assemblage of finds, based on their use and their life cycle.
On the basis of morphological and palaeographical criteria, ten scribes can be identified in Room 7 and twenty-two to twenty-three in Room 8. The total number of scribes identified in studies on the Pylos tablets, in recent years, is estimated to be between thirty-eight and forty-two.Footnote 5 Suffice it to say that this remains a conservative assessment and their true number could be as high as fifty.
The very definition of an ‘archive context’ in Late Bronze Age Greece is not straightforward, as we can see from the work of Olivier and Palaima. Olivier established a framework for the analysis of the administrative organisation in the Mycenaean period.Footnote 6 He recognised five groups of scribal hierarchy, with the archive hall at its highest point, being a place where we have tablets in great numbers, dealing with all aspects of palatial economy, written in situ, or transferred from other areas. There is a variety of Linear B documents (simple recordings, summaries, catalogues), completed by forty scribes or more, but only one archivist. Olivier concluded that Rooms 7 and 8 at the palace of Pylos were the sole example of an archival assemblage surviving from Mycenaean Greece. Later, Palaima went on to define the key features of an archive as this was perceived in the Mycenaean period.Footnote 7 He emphasised that neither the AC nor any other collection of Linear B documents corresponds to a later, more modern, definition of ‘archives’, where we would encounter stored records of historical value transferred for long-term preservation.Footnote 8 Instead, the records found probably cover a period of between two and five months of selected economic activities. In his thesis, Pluta goes on to agree with the criteria and the time horizon set by Palaima for the identification of archive units, repeating once again how they are present in Rooms 7 and 8.Footnote 9 For Bennet, the evidence strongly suggests that this area was the archive of the palace, in which the clay documents were created, managed, and stored to be used as references within the yearly (or shorter) cycles of administration.Footnote 10
Morphological and syntactical features
Usually, the information recorded in each tablet conforms to specific morphological and syntactical rules. The layout of the text makes a clear distinction between the Linear B phonographic signs (syllabograms), arranged at the beginning on the left side of the document, before the ideograms/logograms/sematograms,Footnote 11 representing commodities and numerals, which follow normally on the right side of the tablet, or straight after the descriptive rendering of words in syllabograms. The occasional quantitative entries, on the lower edge of a tablet or sometimes to the side, signify that numerals were important details for both the writers and the readers of these documents, since they cannot be easily memorised. This textual structure occurs on both elongated (leaf-shaped) and page-shaped tablets, with minor variations.Footnote 12 There were neat lines of text (sometimes ruled), with the Linear B signs ordered often in columns. This kind of tabular organisation can be seen mostly in the non-phonographic signs (ideograms/logograms), which lie one under the otherFootnote 13 (Figures 1a, b). The standardisation of script and the apparent attention paid by the scribes to arranging information correctly indicate that they were trying to find the best way to classify readily readable information with as much accuracy and economy of space as possible. This would aid not only the recording process but also the ability of administrators to access the content of these bureaucratic documents easily and efficiently.Footnote 14
Figure 1. Tablets with non-phonographic signs in columns, indicated in black (a, b). (Godart and Sacconi Reference Godart, Sacconi and Serra2019, 350, processed by the author/Killen Reference Killen2024, 336, reproduced with permission of the Licensor through PLSclear).
Textual peculiarities
Having said that, we are obliged to accept certain traits which deviate from our previous description. One of them is the variety in the size of the signs, which is noteworthy. The size of the Linear B characters may differ even within the same tablet. There is no such thing as a common preference or a standardised selection in the size of the signs. It looks as if this matter was not solely dependent on the available writing surface and the volume of recorded information.Footnote 15 Perhaps it was of equal importance for the scribes to present the thematic content clearly and to facilitate the readers’ comprehension of the written text, simply by altering the size of certain signs. Meanwhile, the tabular organisation of the text would allow scribes to obtain all the semantic information needed, while also conducting ‘check operations’.Footnote 16
Linear B texts consist not only of a written language, but also a visual one: with a first, rapid reading of the ideograms all the essential qualitative and quantitative information could be obtained, whereas a second, calmer approach to the syllabograms would offer more detailed descriptions.Footnote 17 The most controversial example of this duality is the so- called ‘double writing’, where words rendered with syllabograms are followed by ideograms of the same meaning. This scribal practice has raised many questions as to its general purpose.Footnote 18 Without omitting any of the evidence, we should point out the obvious difficulty in confirming possible distinctions in all the cases of double writing, while admitting that this practice was employed by both important and lesser scribes.Footnote 19 Whether this system of recording was addressing people of the same or of different levels of literacy, facilitating the reading or offering an alternative perception of the recorded information, remains uncertain. It is possible that the scribes were trying to make their writing intelligible to others, given the small number of literate officials in the Mycenaean kingdoms in general, as demonstrated by many past scholars.Footnote 20 Nevertheless, we cannot help noticing the obvious; that the rudimentary recording and the need to save space, time, and energy would explain the adoption of such a practice, since it allows the easy comprehension of any text inscribed on temporary documents with the minimum effort. In basic terms, the laconic presentation of these records is combined with a commonly accepted textual arrangement to assure a fast, general, yet essential comprehension of what is written.
With regard to scribal practices, our first impression is of a high degree of uniformity.Footnote 21 We do not know exactly how and when this ‘scribal policy’ of uniformity – namely in what way literate state employees applied standardisation and homogeneity within the administrative mechanisms – was set in motion. Nevertheless, it is safe to say that the simplistic recording, which reminds us of rudimentary handnotes, could favour the comprehension and easy adoption of certain writing formulas or scribal tendencies. The interaction between different scribes could reinforce this effort, so enabling the acceptance of general writing traits. The supervision of an archivist or master scribe could also coordinate these parallel procedures, allowing specific scribal patterns to emerge.
The widespread variation of spellings on the tablets would seem to support our hypothesis, implying individual or communal preferences in the scribal production of literate functionaries. Sometimes, individual scribes have a personal preference for the spelling of particular words (as seen with the shift in the use of a2/a, pu2/pu signs or the rendering of /wy/ sequence as wi-j- or u-j-), even though a collective preference for these spellings occurs elsewhere. Tablet writers may also show a constant variation between two or three writing options (as attested by ra2 and ra signs, the Consonant followed by w and Vowel (CwV) sequences and the double R {RR} followed by vowel (RRV) sequences).Footnote 22 Judson emphasises that:
the question of how to spell a particular word could therefore be approached differently at different times, depending not just on the writer's training but also on the word's context, the potential desire to emphasize a particular feature of the word (such as a case-ending) or a whole variety of other, now unreconstructable, aspects of the writer's mindset at that particular timeFootnote 23 … whether in the context of orthography or in other aspects of their writing practices … Mycenaean scribes were practical, flexible writers, employing a range of strategies for the optimum presentation of their administrative documents, and making full use of rather than being constrained by the conventions by which they had learned to use the Linear B writing system.Footnote 24
The obvious lack of hard-and-fast rules which the scribes could use as a kind of manualFootnote 25 does not mean that there are no regulations, but rather that these were not codified or standardised by the palace officials in a strict manner. There is a generalised view of things, as deduced from the great number of recordings of personnel, transactions, and contributions, with no in-depth details about the participants in them. The formation and writing of the tablets, sealings, and labels was mostly the result of a practical spirit. The simplistic schemes and the minimal requirements in terminology used by the scribes remind us of improvisationsFootnote 26 or acts of necessity. It is noteworthy that many clay documents were modified in order to achieve a minimum level of accuracy or, better yet, understanding, every time the scribes wanted to record specific information. Many texts have erasures, re-inscriptions, and additional lines, signs, or words, stacked to a previous record (Figures 2a, b, c). These writing preferences do not seem to expand beyond the scope of better presentation. The visual arrangement and appearance of Linear B signs seem to have been of major importance for understanding a text.Footnote 27
Figure 2. Tablets with erasures, annotated (a, b); or indicated in black (c); re-inscriptions (a, b); and additional lines (a, b). (Palaima Reference Palaima, Duhoux and Davies2011, fig. 12.19, fig. 12.20/Godart and Sacconi Reference Godart, Sacconi and Serra2020, 26, processed by the author).
Having said that, we are confronted once more with the problem of the overall understanding of the various writing strategies employed by the literate administrators over an extended period. Among the purposes of writing administration in Pylos was to present correctly the data that were of interest to the officials, by reducing in parallel both the necessary technical and linguistic vocabulary and the attention paid to standardised grammatical rules. Certain expressions used by the scribes when making specific references to persons and products attest the parallel use of oral communication during the act of writing.Footnote 28 This kind of scribal flexibility could be favoured in demanding working conditions, where the burden of duties calls for immediate solutions. On the other hand, it does not suffice if extended to instances of book-keeping and later use of the textual material, when the scribe is interested in recalling or tracking the transactions and contributions of a specific individual. Even when it comes to double writing, it would not have been easy to write or read the information on a tablet,Footnote 29 without knowing the writing practices of a particular scribe and his idiosyncratic use of signs.
Added to this, the use of various ideograms, enriched by acronyms and monograms appearing solely at one site (e.g. Pylos), poses serious questions as to whether they were understandable to scribes, who were not there at the time these particular signs were written.Footnote 30 The abbreviations in Pylos tablets can refer to different lexemes, depending on the topics covered in the various groups of tabletsFootnote 31 or the individual writing/recording strategies followed by the scribes even within the same centre. The acronym ‘ko’ may represent korwos (or korwā)/boy (or girl), koriandnon/coriander, etc., while the abbreviation O on the elongated tablets of the Sh series has been adopted by Hand 5/605 as a reference to the term o-pa-wo-ta/opaworta/appendages,Footnote 32 in contrast to other scribes dealing with personnel and tax records, where it stands for the term o-pe-ro/opero/debt.Footnote 33 The abbreviation DA refers to persons in the A series and land plots or ‘houses’ in the E series. The occurrence of ideograms/logograms/sematograms without numerals has also been a subject of interest, given that it appears on a more regular basis than expected.Footnote 34 Because of this non-random frequency, it was thought that the information recorded on the tablets was intended to serve a limited number of administrators, making it impossible for someone else, other than the official in charge of the transaction, to figure out the unrecorded numerical data.Footnote 35 Consequently, the idea that these documents were mere aides-mémoire becomes even more convincing.
This administrative reality provides an exceptional link between the written document and the situational context in which it is produced. Even if the attestation of certain ideograms or acronyms at only one site can be the product of accidental preservation of material rather than independent scribal initiative, we should also emphasise their substantial total number, which cannot be coincidental (twenty-eight–twenty-nine iconic/non-iconic signs, twenty-five phonetic abbreviations, three monograms, thirteen determinants).Footnote 36 Under these conditions, one may ask whether all of them were generally understood by most of the scribes. Indirectly, this could indicate once again that written and oral communication among scribes was probably necessary for comprehending the writing practices of one another. As Marazzi emphasises, ‘there was not a single Linear B writing system, but a single notational system adjusted, center by center, writing act by writing act, in as many manifestations of writing as the registration needs required’.Footnote 37 Naturally, one may question whether the non-phonographic use of certain signs could create much (or any) confusion, assuming there was an elementary scribal interaction. Seen differently, however, the creation of new recognisable signs through a sematographic process could impose limitations on the understanding of them by third parties or even by other literate functionaries once a short period of time had passed. Given the nature of the problem, it is surprising not to encounter firm standardisation in vocabulary and grammar, as prerequisite to facilitate the task of every palace official.
The frequency of erasures in Linear B tablets is another noteworthy topic. Several explanations were given in the past, such as errors in recording dictated information, modifications to include additional information,Footnote 38 or hasty decisions.Footnote 39 Logically, the working conditions of the scribes could affect their writing habits and their products. For example, the output of a scribe working in a relaxed situation, of contemplation or solitary writing, will be different from that produced at a time of intense fiscal activity, in which there was a burden of administrative tasks.Footnote 40 It has also been convincingly argued that the Linear B texts were not meant to be read by a broad audience or to be preserved for a long period of time.Footnote 41 For this reason, possible modifications, erasures, or alterations in the texts were not considered shortcomings of the scribes. Many alterations resulted from the updating of written records in the light of new information received.Footnote 42 The overriding interest, however, was the clarity and accuracy of the written information, so that it could be easily retrieved and comprehended.Footnote 43
On present evidence, the work by the scribes can be characterised as rather slapdash, suggesting hasty execution of the tablets. Very often, the signs inscribed are variants or deviations related to their fast or, rather, rudimentary incision. This hasty scribal procedure suggests the casual treatment of the texts by the literate officials. It does not imply a slow, careful processing of transactions, contributions, and obligations, nor are there hints that it could have been to an analytical degree. It resembles more of a general or superficial control of the activities involved. It certainly bears little resemblance to the meticulous writing forms we would expect for an archival assemblage aimed at listing personnel, products, and services in as much detail as possible. Additionally, it is reasonable to suppose that the clarity of the signs would have been a sine qua non demanded of their inscription. The accurate and easy perception of the content in Linear B texts would definitely be a precondition for third parties or even for the master scribe. This scribal principle is far from the reality attested in Pylos. To be fair, the crude treatment of inscribed signs on the Pylos tablets and the visual result would probably be difficult to deal with in an archive deposit, after the elapse of a short period. For the page-shaped tablets in particular, their content produced by several scribes other than Hand 1, would require meticulous treatment – as time passed and new additions were made to the documents – to verify, adjust, and specify former scribal procedures. Later interventions would make it more difficult to synchronise and keep up with the same or supplementary administrative events, even if there was just one person responsible for the ‘literate updating’ of an archive (e.g. a master scribe/archivist).
Other morphological traits point to the slipshod and rudimentary preparation of the tablets. The blank or incomplete lines/entries, the hastily incised and irregular lines, the surplus of empty space, usually at the lower end of the Linear B records, the frequent cases of Linear B finds with signs drawn on their back (verso), as well as the existence of signs on the right and left sides of clay tablets, due to cramped writing or possible modification, testify to a ‘pragmatic’ rather than far-sighted approach to the writing medium, even when page- and card-shaped tablets are concerned (Figures 3a, b, c, d, e). Each of these traits individually may not be sufficient to be regarded as a scribal peculiarity. However, when seen all together, the picture of an unorthodox, ad hoc, and shortsighted use of the writing space becomes more obvious. We should note that these morphological and textual peculiarities are more common than we think. For example, they appear in one form or another on more than eighty page-shaped tablets out of a total of roughly 140. In fact, they are associated with the output of many Pylian scribes (Hands 2/602, 21/621, 24/624 etc.), including Hand 1/601. These ‘clumsy’ features can be seen in many tablet series, with texts recording personnel, land allotments, commodities, and other raw materials. If we take into consideration the patchy surface of many documents containing compilations and summaries, along with the multiple (textual, formatting, syntactic) editing observed on them, we may doubt their value as archive material for anyone other than an official in charge of and knowledgeable of these written sources, such as Hand 1/601. Even then, the chief scribe would be incommoded by the management of such ‘peculiarities’ when observed in other scribes. If we suppose that this was a recurring phenomenon, then the volume of these rudimentary, incomplete, hastily written, and variously edited tablets would pose a serious problem regarding their validation and classification by a third party, within a burdened working environment.
Figure 3. Tablets with blank lines/entries (a); hastily drawn irregular lines (b); empty space (c); signs on the back surface (verso) (d); and on the right and left sides of clay tablets (e). (Godart and Sacconi Reference Godart, Sacconi and Serra2019, 352, annotation by the author/Godart and Sacconi Reference Godart, Sacconi and Serra2020, 38, processed by the author/Killen Reference Killen2024, 345, 396, 398, reproduced with permission of the Licensor through PLSclear).
In addition to the morphological traits, remarkable heterogeneities occur with the syntactical traits of Linear B documents. For example, the text on the leaf-shaped tablets may fill the whole width of the writing surface, but it could also cover just the lower part of the tablet (Figures 4a, b). Sometimes the same scribe may use both practices, like Hand 23/623 in the Ad series. The use of ruled lines is not a steady principle among scribes (Figures 4c, d). They may apply it, but it remains optional. Cases of lines extending to a certain length but not covering the whole surface of the clay document are also observed. Scribes may draw lines on the tablets in one series, but prefer not to in another series, like Hand 1/601 in the Ae (without lines) and the An series (with lines). Other times, we see the same scribe using these syntactical traits simultaneously in the same series of documents, as in the case of Hand 41/641 in the Eo series, where we have ruled and not ruled palm-leaf-shaped tablets. As a general observation, the use of tablets without ruled lines is the commoner practice among scribes, even the ones who seem important (e.g. Hands 1/601 and 2/602). Ruled or partially ruled clay documents also appear, although these seem to be adopted solely by a distinct group of scribes with significant production (Hands 1/601, 2/602, 21/621, 41/641).
Figure 4. Text placed in the central part (a) or the lower part (b) of the tablet. (Godart and Sacconi Reference Godart, Sacconi and Serra2019, 28/Killen Reference Killen2024, 324, reproduced with permission of the Licensor through PLSclear). Lines covering part of the tablet (c) or the whole surface (d). (Godart and Sacconi Reference Godart, Sacconi and Serra2019, 32/Killen Reference Killen2024, 357, reproduced with permission of the Licensor through PLSclear). Different patterns of text arrangement (e, f, g). Syllabograms indicated in red, ideograms and numerals indicated in blue (Godart and Sacconi Reference Godart, Sacconi and Serra2020, 61, 216, processed by the author/Killen Reference Killen2024, 351, reproduced with permission of the Licensor through PLSclear).
In parallel, the deployment of textual information follows differentiated patterns: usually the syllabograms come first, with the ideograms and the numerals placed at the end, on the right side of the clay document (Figure 4e). This standardised arrangement may change from time to time, becoming more complex: the A-series tablets and the Ma series present the standardised text deployment, completed with more abbreviations or ideograms, followed by numerals in a specific order (Figure 4f). Many scribes adopt this method of registration (Hands 1/601, 2/602, 4/604, 21/621, 23/623, 51/651). On the opposite side, some literate functionaries prefer to simplify matters: in the Ua and Ub series the scribes write down just the syllabograms or just the ideograms/logograms, followed by numerals (Figure 4g), a practice preferred by many scribes (Hands 14/614, 31/631, 32/632, 42/642).
The emphasis on the non-phonographic signs, the sematographic part of each tablet, is another distinctive characteristic of the literate officials. In many cases the ideograms/logograms and numerals are written at a distance from the rest of the text, in a scribal attempt to focus attention on them through the spatial organisation of the document (Figure 5a). The position of the non-phonographic signs in columns supplements this writing strategy, to produce a better visual result (Figure 5b). Various scribes may resort to these methods, although it is not a phenomenon attested for all of them. In relation to the need to stress the importance of certain signs, scribes frequently prefer to increase the size of the proper ideograms/logograms or syllabograms, so as to emphasise the particular information which was of interest to them. In the Ea and Eb series, Hands 41/641 and 43/643 enlarge either the ideograms or the syllabograms (Figures 5c, d), while in the Ae series the same happens with the initial syllabograms on the left side of the tablets by Hands 1/601, 13/613, 22/622, 42/642. The obvious aim behind this scribal mechanism is to address different clues in every Linear B record, which are not of the same importance to every scribe.
Figure 5. Ideograms/logograms and numerals written at a distance (a) and in columns (b). (Killen Reference Killen2024, 351, 363, reproduced with permission of the Licensor through PLSclear). Larger ideograms (c) or syllabograms (d) inscribed on the tablets. Syllabograms indicated in red, ideograms and numerals indicated in blue (Godart and Sacconi Reference Godart, Sacconi and Serra2019, 221, 239, processed by the author).
The page-shaped tablets may also present non-homogeneous syntactical features. Beginning with the text arrangement, the norm here includes single or double columns with ideograms/logograms and numerals, preceded by phonographic signs (syllabograms). This kind of deployment is attested in different series of documents, such as records of personnel, landholdings, food rations, distributions in metal, tax payments, etc. Many scribes turn to this writing scheme, while trying to keep a clear distinction between the columns of ideograms/numerals and the syllabograms (Figure 6a). Other times, this tabular arrangement does not exist or is unclear (Figure 6b). The lack of a columned structure can be seen in more than one series of tablets. There are examples of scribes (Hands 1/601, 3/603, 24/624, 42/642) who follow both writing habits. Another syntactical feature is the use of blank lines to distinguish between separate paragraphs or to make the document more conspicuous (Figure 6c). The use of paragraphs was adopted by many scribes (Hands 1/601, 2/602, 3/603, 24/624, 63/663, etc.), without of course being accepted as a general writing principle by all of them (Figure 6d). Last, we encounter unusual syntactical features: in PY Tn 316 the size of syllabograms pu-ro (Πύλος) is increased at the beginning of every paragraph (Figure 6e). This uncommon habit constitutes another interesting scribal peculiarity, which is without precedent in the Pylos tablets.
Figure 6. Tablets with tabular arrangement (a); without distinct tabular arrangement (b); with separate paragraphs (c); without paragraphs (d); PY Tn 316 (e). (Godart and Sacconi Reference Godart, Sacconi and Serra2019, 129/Killen Reference Killen2024, 337, 342, 374, 397, reproduced with permission of the Licensor through PLSclear).
Turning to the size of the signs inscribed on the page-shaped tablets, we observe considerable variations, since Linear B signs range roughly from 0.5 to 1.5 cm in height. The fluctuation in the size of signs is characteristic of many clay documents written by different scribes, some of them regarded by scholars as important (Hands 1, 2, 21 among them). This kind of alteration within every text reflects hastiness or indifference in execution, which could be related to the temporary and expendable nature of the clay record. Generally speaking, the spatial deployment of the texts is not as standardised as we would expect. It remains doubtful whether there was an intentional centralised/administrative guideline behind their formation. This is a fact worth considering, should we assume that even a repository of documents which rotate on regular one-year(?) cycles would have gradually developed an adequate scribal homogeneity for practical reasons. Combined with other characteristics of these tablets (variable size of signs and textual arrangement adopted by different officials, incomplete/rudimentary content, erasures/editing), these counsel caution in accepting them as ‘archival’ documents, in the broad sense.
Overall, the text layout on both palm-leaf and page-shaped tablets may conform to embedded scribal conventions, capable of ensuring the effective arrangement of information and the subsequent understanding of writing and administrative techniques. Nevertheless, the fact remains that there is no consistent use of these techniques by the scribes, in a manner which would allow us to connect them to well-defined systematic, morphological, and syntactical patterns. Clearly, formality in morphological and syntactical traits, through the general acceptance of common scribal rules, is absent in this ‘official’ collection of tablets. There is a non-obligatory set of rules governing the layout of the texts. It seems that the scribes of the tablets were following their own preferences in text arrangement, without being obliged to comply with rules of scribal uniformity, as we would expect in a more organised collection of documents used for archival purposes. Literate officials working inside and outside the palace could produce a significant number of Linear B tablets every day, which were then transferred to the AC in Pylos. As time progressed, the intentional lack of control over the writing procedures and the attested scribal idiosyncrasies would result in problems with managing a centralised assemblage of clay documents. Meanwhile the effort and time needed for this system to be thoroughly understood, primarily by the master scribe and secondarily by any literate administrator of the palace, would have become even greater. In terms of practicality, standardisation would have been a firm requirement, whether it was short-term or long-term recordkeeping that was taking place.
Some researchers claim that the overall effort made by the administrators to maintain the ongoing validity of their documents through erasures, re-inscriptions, and syntactical modifications can be taken as an argument against the widely held view that there was a further stage of recording and storing information on a perishable medium.Footnote 44 For some researchers, the care taken in correcting and editing clay tablets, again and again, proves that these were significant documents worth considering as such by the Pylian scribes. Yet, under these circumstances, we may also argue that there was an overarching need to achieve correctness at all levels before the final transcription of all the information onto another writing medium (wood, papyrus, leather) that is not preserved in the archaeological record. We should add here that the signs of Linear B consist of curved as well as straight lines, more suited to writing or painting with ink on papyrus, animal skin or wood than to incising on clay.Footnote 45 Furthermore, the simplistic recording followed by the scribes, even in the page-shaped tablets, resembles that of personal handnotes, which could always be due to the oral transfer of specific information from one functionary to another. The repetition of this rudimentary process many times for every tablet raises the question as to whether this scribal tactic could produce a credible corpus of files worth saving and storing for an extended period.
Writing material and pinacological data
The reasons Linear B scribes chose specific writing materials are due to their effort to obtain the appropriate medium for recording, classifying, and transferring vital information through the writing process. The use of clay as a writing surface is certainly one of the most logical decisions to make, judging by the supply available to the Mycenaeans from their surrounding environment. One of the main aims of scribes and related officials was to save time, material, and energy during the administrative practice of recording selected transactions. Clay is available in large quantities and can be processed by specialist craftsmen or less specialised workers to produce writing documents (tablets, sealings, labels).
Examination of the material of these documents revealed differentiations in their clay fabric, implying variations in the materials used, their place of origin, or mode of production.Footnote 46 The complicated picture that arises with heterogeneous clay fabric could be due to incoming tablets, brought from places outside the palace. Although a rational proposal, the idea of tablets assembled from other settlements in Pylos is still controversial. Pape has proposed that rehydrated clay documents, especially palm-leaf-shaped tablets, could be re-inscribed a long time after their initial production.Footnote 47 If we accept that tablet-processing could take place at any given moment, this does not mean they were originally made in situ outside the palace. Scribes could use other more portable documents, for example sealings, to register objects or transactions in settlements all over the kingdom.Footnote 48 Tablets could have been prone to accidents, due to transportation over rough terrain, unpredictable weather conditions (rain, moisture), and the physical restrictions of the material itself, such as exposure in a non-homogeneous drying environment, which would result in cracks and breaks. Covering short distances with a bulk of elongated tablets is possible,Footnote 49 but it would also be much easier to make these clay objects from a clay source, or sources, close to the palace, allowing provincial rulers and officials to turn to more effective and sustainable recording means (e.g. sealings).
It is logical to assume that the centralised administrative model, recognised in many forms at the palace of Pylos (judging from its architectural layout, the distribution of finds associated with administration, and finished products), would have extended to certain aspects of scribal organisation and planning. In all cases, it would have been in the palace's interest to secure a steady supply of clay from a particular source or even a uniform production pattern for all the tablets of the AC. Unlike the tablets from the eastern Mediterranean, which were often baked before the filing and storage procedure, the clay documents from Pylos were sun dried and have been preserved by chance, due to the conflagration of the palace at the end of the thirteenth century BCE. The use of sun-dried clay would be adequate testimony ipso facto of the temporary character of the Pylos documents. What is controversial, is the lack of correlation between these deposits of tablets and the archival nature attributed to them. It is safe to say at least that the heterogeneity of inclusions in the clay fabric of the tablets points to an indifference regarding their quality or to lack of a standardised method of production.Footnote 50 In terms of technical features, the facility in using and processing clay permits the use of the material under any working conditions. That also meant clay was one of the first writing materials the scribes would turn to, not just in normal circumstances but also in the face of any extreme and non-predictable events. This choice would become the preferred option in times of large-scale recordings of products and transactions or in relation to administrative practices demanding constant scribal registration of personnel, materials, and services.
The physical forms of the Linear B tablets also present considerable variations. The shapes, sizes, and thicknesses of the tablets are so diverseFootnote 51 that the contradiction of the notion of uniformity is remarkable. Any difference between the preserved palm-leaf, card, or page-shaped tablets, or even between tablets of the same category, militates against the homogeneity usually preferred in administrative practices. The lack of standardisation in either their fabric or their shape and the apparent lack of interest in achieving uniformity would be an unexpected turn in an organised recording procedure, especially one aimed at the renewal or creation of a collection of documents, that was intended to be stored over a number of months. The development of defined characteristics in shape and form would have been a logical choice, which is totally absent from this so-called archival assemblage.Footnote 52 Instead, tablet-makers were left to decide about the basic features of the Linear B documents (dimensions, general shape, clay fabric), based on individual considerations which elude us. Scribes do not seem to have developed distinct preferences for specific types and sizes of documents, regardless of the administrative or practical needs dictating these choices.
Depending on their size, the tablets’ weight can differ significantly. The author's examination of fifty-five page-shaped tablets from Pylos (stored in the National Archaeological Museum of Athens) showed that their weight ranges from 67 gr. to 916 gr. (Table 1).Footnote 53 Due to the fragmentary condition of many finds, it is logical to expect their normal weight to range from at least 100 gr. to 1000 gr. or more. The number of documents weighed constitute about 40% of all the page-shaped tablets recovered from Pylos, enabling the extraction of well-justified conclusions based on a reliable set of data.
Table 1 Weights of page-shaped tablets (created by the author)
The figures presented below corroborate our views on diverse techniques of fabrication, as well as an obvious tolerance for this practice, accepted even by the high-ranking officials housed in the palace. Altogether, the current data suggest that tablet production was a more complex and less centralised affair than previously considered.Footnote 54 Rolling or folding pieces of clay were two techniques used by the tablet-makers, but with no special preferences associated with the type and size of the tablets, or even their thematic content. Even the use of straw or strings in the clay was an option not associated with specific scribes or series of Linear B documents, while its adoption for administrative purposes is still debated.Footnote 55 Based on the research conducted so far, the making of tablets seems to have been a highly individual process, with no clear connection to the administrative environment, leading us to doubt the idea of tablet production controlled and managed by state functionaries. Furthermore, we begin to wonder whether the tablets’ various fabrics, shapes, and sizes could facilitate their filing and categorisation by a third person (a master scribe?) after a period of a few weeks.
Complementary to the type of tablets used, the ratio of palm-leaf to page-shaped tablets in the AC is of equal importance. Only one-fifth of all written documents in Room 8 are page-shaped tablets, while the number of this type of documents in Room 7 is smaller (roughly one-tenth of all preserved tablets). Given that page-shaped tablets are used mostly for summaries and compilations, we would expect a higher percentage of collective documents of this kind inside the AC. Their low frequency among the prevailing palm-leaf-shaped tablets is of little help in referring to the existence of a permanent archive unit, in relation to the Linear B tablets found elsewhere in the palace. After all, preliminary palm-leaf tablets discovered in several sectors of the palace are the norm, never the exception. It could be argued that their preservation in an archive along with summaries could serve the needs of the palace scribes to verify older recordings at any given moment through preliminary documents. That said, the concentration of a substantial number of documents in this area reveals much about the function of Rooms 7 and 8 as building sectors dedicated to scribal activities, but it does not presuppose their constant use for such a purpose or the duration of the recordings taking place in situ.
Physical context and its relationship with the Linear B finds
Written documents have three partly overlapping functions: as means of communication, as aid to memory, and as evidence.Footnote 56 The organisation of any archive for future recourse is a carefully planned process which should take this reality into consideration. Consequently, it demands a proper exploitation of space, a thorough assessment of the classification of the tablets, and a widely accepted understanding of its function by third parties, in case of need or emergency. There is nothing clearly showing that these factors were correctly evaluated or applied in the so-called Archive Complex in Pylos.
Re-examining space exploitation
Beginning with the exploitation of space, it is easily understandable that record-keeping of any kind could not be done solely inside Rooms 7 and 8, which were too small to be used for the systematic arrangement of tablets (Figure 7). Room 7 was a small square area (4×4 m). Among its finds were animal bones and twenty-two diminutive kylikes in the west part, a spearhead, and a sword.Footnote 57 In its south part was a large pithos, probably filled with water, which was necessary for making new tablets or destroying those that were no longer needed.Footnote 58 The placement of the vessel would significantly reduce the available workspace for the tablet-makers or the scribes who were involved in producing the clay documents. The humble size of the adjacent Room 8 (roughly 4×3 m) and the further restriction of the workspace in it, due to the existence of a bench on three sides, wooden shelves, and a (possible) second doorway in the north-east wall,Footnote 59 towards Area 2, are also crucial factors, which exclude any attempt to sustain a massive collection of clay documents. Neither of these two rooms had evidence of floor or wall plaster fallen from above. There are no artifacts whatsoever that suggest an upper story existed over the AC, although the destruction of the surrounding walls had led to the formation of badly disturbed areas.Footnote 60 Nevertheless, we would expect some evidence for a floor structure to have been preserved. The lack of upper stories would further deprive the building complex of spaces necessary for producing and storing tablets, making it impossible for an archivist or master scribe to operate in a functional and effective way, given the limited working environment available to him. It is even more difficult to accept such a scenario, if we consider the nearly 200 tablets found elsewhere in the palace, outside the AC, as administrative documents not intended to remain in the nearby areas but waiting to be collected in Rooms 7 and 8Footnote 61 (a procedure unfortunately interrupted by the destruction of the palace). If we also bear in mind the extensive production of Linear B tablets, which could have happened daily, the systematic filing of Linear B documents in large numbers would have been impossible.
Figure 7. The Archives Complex (Palaima Reference Palaima and Brosius2003: fig. 8.2).
Judging by the tablets’ distribution patternFootnote 62 (Figure 8), the north-east wall of Room 7 could perhaps be used for storing purposes. Still, the circulation of people along this side and between two doorways, towards Area 1 and Room 8, seems to advocate an alternative interpretation of their deposit. Any concentration of written documents on shelves following this axis would hinder the movement of persons from one room to the next. As the wall on this side no longer exists, we cannot speculate over the position and the size of the entrance.Footnote 63 The lack of burnt wooden traces or hinges, like the ones recovered from Room 8, argues against the possibility of shelving and the overall picture presented by the archaeological data provides little evidence for storing practices. Other ideas, such as the temporary placement of Linear B documents in baskets or inside containers of perishable material, on the floor, close to the corner space between Rooms 7 and 8, could explain much more easily the general distribution of the finds. There is no reason to exclude the possibility of a surplus of documents, waiting to be registered or simply transported to the neighbouring Room 8, just as was proposed for the Linear B tablets and sealings in the so-called grid 52, in Room 7.Footnote 64 In a proposed reconstruction of the north-east wall, which was replaced by the Chasm, Palaima proposed that there was a depression, about 50 cm wide, which could correspond to a shelf or a table, projecting from the wall.Footnote 65 He correlated it with the width of the north-west bench in Room 8, proposing similar shelving arrangements for Room 7. It is true that most tablets in the grids close to the wall are neatly grouped according to their series (Ad, Ea, Sa, Ta). The existence of another shelf was suggested in grid 83, near the south part of the room, towards its entrance.Footnote 66
Figure 8. Distribution of tablet fragments in the AC (Pluta Reference Pluta1996: fig. 7).
However, this interpretation does not necessarily reflect the real picture of the now- perished fixtures or other structures and furniture of this room. Palaima's space- restoration hypothesis remains an argumentum ex silentio, with little theoretical evidence to support it. The whole idea of shelves contradicts the tablets’ context, since the grouping pattern that emerged from the finds is unlikely to have resulted from storage on different, continuous shelves, one above the other. Instead, we would have a more differential pattern, due to the collapse of the shelves and the scattering of the finds in mixed assemblages. Even the existence of a shelf or a table close to the lowest part of the wall can be refuted because of the possibility of a bench or a table on the west side of the room, where there was potentially more free space to use it, by analogy with a similar structure in Room 8 (clay bench). Unfortunately, the absence of Linear B finds in abundance close to the north-west party wall (between Rooms 7 and 8) makes it more difficult to accept the existence of shelves in this part of the room. Nothing can be said for certain about Palaima's spatial arrangement of activities along the north-east wall. It is one of the two proposals that have prevailed so far, with the other, which is even simpler to my mind and would seem to make more sense, being the placement of the clay tablets in perishable containers (wicker baskets?) left temporarily on the floor to be registered or collected by the scribes.
Re-examining classification patterns
Pylian documents were categorised following patterns that are not so well known or not readily related to one another. The Sh, Ta, and Ma series had received some attention in the past, because of their grouping practices, which present interesting differentiation.
In the Sh series PalaimaFootnote 67 has identified the existence of contiguities, used to process tablets as a set of texts. The palm-leaf tablets are receipts or tokens for renovated sets of defensive armour. In the Ta series we have a group of thirteen tablets containing descriptions of over seventy objects collected for use at a sacrificial banquet, to mark the royal appointment of a man as da-mo-ko-ro. The Ma series lists an assessment of six commodities, not all of which we can safely identify but which appear to be in a fixed ratio to one another, with the absolute quantities varying by district.Footnote 68 The seventeen Pylos Ma-series tablets correspond to one tablet for each district of the kingdom, separated into the Hither and Further Province. Their dispersal in Room 8 shows that the tablets from the Hither Province do not overlap with those from the Further Province.Footnote 69 It is reasonable to assume that the scribes kept the tablets from the two provinces in separate spaces and that they were scattered by accident, although some other explanation cannot be excluded. Having said that, their classification seems to follow a more composite clustering, related to fiscal and economic practicesFootnote 70. In the Jn seriesFootnote 71 it is unclear whether the location of the smiths or the exact number of them working (or not working) under the ta-ra-si-ja system is the most significant recording noted by the scribes. Similar thoughts are prompted regarding the functionaries from the Jn and Jo series (ko-re-te-re, po-ro-ko-re-te-re), asked to provide specific quantities of metals from every region of the kingdom. Again, there is no way of knowing what was of greater importance to the palace: the place or the products sent by the local officials and their consequent identification. Any of these parameters would easily play a crucial role in the classification and filing procedures followed by the scribes in Pylos.
There are also the labels on baskets with tablets. The text on the labels provided a summary for the contents of the basket to which they were attached. The information inscribed could serve as a guideline in the final phase of filing the tablets on shelves.Footnote 72 Thus, the label affixed to the outside of the container facilitated sorting procedures, once the group of Linear B records reached its destination.
Other administrative documents saved in the so-called AC are the sealings. At Pylos, there are 165 sealings, with twenty-three of them being inscribed. Only one inscribed sealing was found in the AC. The purpose of the inscription was to transmit basic information to the administrators about the items, to which the sealings were attached.Footnote 73 Quite often, the impression of a stamp, without any extra inscription, provided all the necessary information. Sealings of this type would therefore function primarily as recording instruments within transactions that do not require the use of writing.Footnote 74 Their use could also extend to prevent tampering with closed wooden boxes, which contained tablets, in Room 8.Footnote 75
All these finds point to different sets of documents, organised according to indiscernible criteria, to meet the needs of the officials working in Rooms 7 and 8. The fragmented condition and dispersion of most of the Linear B tablets in Room 8 does not permit us to identify clear grouping patterns, thus limiting our ability to verify the bold claim that this is the basic storage and filing area of an archival complex. The sense of record-keeping in Rooms 7 and 8 lasting for a limited time period is further attested by the pinacological and textual evidence: clay tablets were left unbaked so that they could be recycled and reused, once they had served their administrative purpose, while sealings functioned as nothing more than temporary notes for later tablet work. The inscriptions on sealings are not ‘official’ enough to argue for the prolonged usefulness of these documents, nor should such a function be attributed to such small, fragile items. In parallel, most of the uninscribed sealings were carelessly made and found broken and discarded, suggesting they were intended to have a short lifespan.Footnote 76
Another hint pointing to the limited life cycle of Linear B tablets is the absence of time references, which are present in only a small number of documents. The idea of a maximum one-year consultation periodFootnote 77 was due to the study of taxation records, especially the Ma series (eighteen tablets), in which are attested the terms pe-ru-si-nu-wo/wa (last year), za-we-te (this year), and a2-te-ro we-to (next year), in correspondence to the delivery of six particular products, acquired from the sixteen geographical districts of the kingdom. However, we should emphasise that time references of this type occur on only eight of the eighteen Ma tablets. Therefore, we can merely speculate about the life cycle of specific Linear B documents out of the whole series. In addition, these eight tablets are related to tax revenues coming mostly from the Further Province or some remote areas of the Hither Province. The probability of the palace at Pylos exercising control over the adjacent region, but having less direct and detailed control over the economy in distant geographical districts, was proposed many years ago after careful examination of the land holdings in the E series.Footnote 78 In addition, researchers have argued in favour of a three-tier hierarchical system in the Mycenaean administration, with the palace on top and second-order centres or lesser settlements scattered all over the countryside.Footnote 79 These minor centres were under the control of local officials and institutions, responsible for the collection of taxed commodities, which were of interest for the palace.Footnote 80 It is thus logical to assume that the provincial districts could monitor and record the acquisition of products and raw materials, either through oral communication or with the aid of written media (seals, sealings, tablets). In any case, it would not require the prolonged safekeeping of Linear B texts in Pylos for a year, on the condition that any information about tax payments could easily be channelled towards the palace when it was needed. Consequently, the discovery of clay documents related to tax payments, in Rooms 7 and 8, does not presuppose their retention for an extensive period of time. Even the ‘debts’ (o-pe-ro) owed to the palace from the previous year may have already been discharged when the scribes were in the process of inscribing the current payments coming to Pylos.Footnote 81 After that, discarding the clay documents would be the logical choice to make.
Even if we accept a scenario where some tablets referring to yearly activities were indeed stored in the palace, for a period of up to a year, we end up with a total of twelve tablets (eight from the Ma series, two from the Ub series, plus Aq 64 and Es 644), where similar (annual) terms are attested. Moreover, we realise we know nothing about their final storage place or their distribution in space. We have one or more deposits of a few tablets, with no further hints to relate them to a substantial collection of administrative documents. By itself, this insignificant quantity of recordings cannot be associated with larger sets of documents. If stored in the palace, their integration in a broader corpus of written files cannot be justified by the remaining scarce evidence.Footnote 82
So far, the debate about the way the Pylian scribes classified their documents has proved inconclusive. Perhaps, a generalised criterion for the broader classification of written documents was the place name or toponym, where a person is witnessed or a transaction is made. Another possibility is the classifying of named individuals for demanding administrative purposes.Footnote 83 For example, the recording of 200 smiths by name could have been related to their responsibility for producing finished goods from metal supplied by the palace. The total number of named individuals recorded in the Pylos texts is approximately 700, from all over the kingdom.Footnote 84 These 700 persons represent on average 20% of the personnel associated with the palace and roughly 6% of the adult male population in the kingdom.Footnote 85 Named smiths and herders range widely in social status.Footnote 86 Therefore, their scribal attestation could be solely dependent on their specific administrative duties. Bearing this in mind, we tend to accept that any classification patterns that existed in Rooms 7 and 8 were formed based on principles, perhaps subjective, defined by the scribes or the master scribe working there. Sometimes, the ever-changing administrative needs could set the principles for any necessary grouping of Linear B texts. To this end, it is reasonable to imagine that functionaries took notes on oral suggestions as to how they could organise their material or modify it over time. These oral guidelines could play a crucial role in producing, renewing and consequently stacking the Linear B records.Footnote 87
Since we know so little about the administrative approaches which defined the filing arrangement of the AC, we are also left wondering whether the archivist or master scribe was ever in need of a conscious and constant effort to deal systematically with the textual documents at all. Considering the tablet output of the total number of twenty–twenty-three scribes, who were dispatching their records to the AC, and the limited capacity of this space, there is reasonable doubt as to whether this volume of Linear B documents could be kept intact for a long period and, if so, whether their content was considered valuable enough to be organised in a systematic, scrutinised way. Complementary to this opinion is the issue of the low percentage of page-shaped tablets in Room 8. Only one-fifth of all written documents are page-shaped tablets. Their low number in relation to the preliminary palm-leaf-shaped tablets inhibits us from acknowledging the existence of a rather credible and steady collection of documents. Keeping a more generic perspective, we end up with an astonishing assemblage of documents lacking obvious classification and filing patterns not only in form (e.g., low percentage of page-shaped tablets) but also in content. We are thus left with strong doubts about the consultative character of the records kept in Rooms 7 and 8, as well as the potential of them being viewed as a safe, future source of information by the local scribes.
Re-examining the status of scribes and issues of collaboration
When considering the evidence for the functioning of an organised archive complex, we must be cautious as to whether it was considered as such by the scribes themselves or by third parties. The presence of an archivist or a master scribe points to the existence of scribal hierarchy among the literate administrators. Hand 1 is responsible for the production of about 237 tablets, while he dealt also with the editing and correction of several clay documents written by other scribes. We should not forget, however, that other scribes of high administrative status were also in Pylos. Hand 2 is also a prolific writer, who seems to deal with economic issues carried out within the palace.Footnote 88 Both Hands 1 and 2 have inscribed many tablets, in comparison to other scribes, which refer to a wider variety of subjects. At the same time, their bureaucratic responsibilities could differ.Footnote 89 Hand 2 is more ubiquitous as a scribe than Hand 1, given the fact that his tablets were found in many rooms and the AC, while the work of Hand 1 is restricted exclusively to the AC. It is also evident they both co-operate with different scribes in different areas of the palace. Consequently, the relationship between scribes, such as Hands 1 and 2, raises serious doubts about the potential of having a single scribal summit at the top of the Mycenaean administration at Pylos. This could lead to further implications when it comes to archival matters, regarding the responsibilities for the organisation of the corpus of Linear B documents and the degree of cooperation among peer scribes.
Moreover, the tablet deposits of the palace may have been different depending on the time of year the recorded transactions took place. Not only the number, but also the distribution of Linear B finds could vary in size and context.Footnote 90 Similarly, the scribes involved with the literate administration of the palace may have been different persons from time to time. The eventuality of scribes being under-represented or singled out, depending on their overall duties during the year, is not improbable.Footnote 91 The latter follows Pluta's suggestion that ‘it seems unlikely that so many scribes would be put through the rigor of learning how to write in the service of Mycenaean administration only to write a couple of tablets per year’.Footnote 92 There is nothing to attest the permanent employment of literate administrators for specific duties, mainly because the picture provided by the archaeological data refers to the time immediately before the destruction of the palace. Many of the surviving texts appear to have been inscribed by a relatively small number of literate individuals. According to some researchers, the authors of the Linear B documents, typically referred to as ‘scribes’, were in fact literate administrators for whom writing was part of their professional skills and activities.Footnote 93 These administrators were dealing with large numbers of individuals, as seen by the 700 complete names that have survived on tablets, while also fulfilling other social and economic duties. They could have been part-time scribes, not necessarily assigned to a stable work post. We are thus faced with the fact that these ‘seasonal’ scribes would have had serious trouble handling a pre-existing archive, arranged in ways not known to most of them. Moreover, the non-standardised writing conventions would further confuse any attempt made to emulate previous recordings or scribes.
On the other hand, even if there was a permanent master scribe or archivist in the AC, he would not have been able to manage on his own an ever-increasing corpus of Linear B documents, diverse in syntax and content, accumulated in a short period of time by as many as fifty scribes working in Pylos. Approximately 200 out of the 700+ tablets of the AC were found in Room 7 and another 500 documents were recovered from Room 8 and its vicinity, towards Area 2. These numbers should be juxtaposed with the augmented scribal production expected to result from forty–fifty literate officials and the limited storage capacity existing in Room 8 (where most of the filing and record-keeping activities were taking place). In total, the time needed for an experienced official or scribe to form and fill in a palm-leaf-shaped tablet would be roughly ten–fifteen minutes, while the card- and page-shaped tablets would take a little longer, due to their multiple lines and entries (about fifteen–twenty minutes). In this hypothetical procedure, an experienced literate official/scribe (e.g. Hand 1/601) would produce four–six elongated tablets or three–four page-shaped tablets per hour. Thus, he could record, in relative ease, a maximum of sixty leaf-shaped tablets or forty page-shaped tablets per day.Footnote 94 The number of scribal Hands recognised on the basis of morphological and paleographical criteria in the AC ranges from twenty-six to twenty-nine. Based on the figures given above, we may speculate that ten persons would be responsible for the production of 600 leaf-shaped tablets or 400 page-shaped tablets per day. Should we counter-argue that these were inferior scribes, or engaged in other duties, then their tablet production could potentially be half that of Hand 1 (300 palm-leaf-shaped tablets or 200 page-shaped tablets per day). If we further reduce their productivity, to a quarter of that achieved by Hand 1, or even limit the number of scribes to half (five), we will then have 150 palm-leaf-shaped tablets or one hundred page-shaped tablets per day.Footnote 95 After careful estimates, one instantly realises that even a minimum literate production accomplished by these scribes would leave no space for a sufficient assemblage of documents reaching considerably back in time, due to the easily exhausted record-keeping capacity inside Room 8, within a few days or weeks. In addition, the replacement of the ‘seasonal’ scribes and their use for different assignments throughout the kingdom would further complicate any attempt to process and organise the writing material, without any proper assistance by the tablet-writers themselves.
Recently, it has been suggested that a significant number of similar names from different series of the Pylos tablets belong to the same individuals,Footnote 96 as in the case of twenty-seven smiths, who also seem to be herders, or the four–five smiths, who are associated with the goddess Potnia, or the nine smiths appearing in the o-ka tablets as military personnel of a certain rank (officers and administrative elite), and the sixteen smiths recurring in landholding documents. Some herders, on the other hand, are recognised as craftsmen, landholders in the E- series, and are possibly referred to in the M- and N-series for tax payments. NakassisFootnote 97 has proposed that these persons constitute a group of important multi-tasking agents, a broader regional elite which could verify and assist any scribal undertaking from the palace of Pylos. Even so, this rudimentary recording of the named individuals would not facilitate their identification as persons or the confirmation of their working and economic activities over a long-term period, when Pylian scribes were responsible for the closure of numerous daily transactions. Many important individuals are often mentioned just once, thus increasing the uncertainty over their secure identification. Possible scribal interactions, especially in the form of oral communication among the literate personnel, could resolve some bureaucratic issues, but not the ones extending over a long period of time, as in the case of an archival collection.Footnote 98 Given the quantity and variety of transactions recorded each day, it would have been impossible to keep track, or even have a recollection, of persons and activities of the preceding few months, while the rapidly exhausted storage capacity in Rooms 7 and 8 would have forced the officials to discard a lot of the writing material which was already collected.
A final point we should address is the likelihood of having a class of professional scribes working for the palace. Hands 1 and 2 are responsible for an impressive number of tablets in several areas of the state economy, but they are also employed in correcting and re-writing the work of other scribes. Given the evidence of their firm presence in the so-called AC, it may be suggested that they are valued persons of high status at Pylos.Footnote 99 Shelmerdine argues for the possibility of professional scribes at PylosFootnote 100 (eleven non-specialised, as opposed to twelve fully and semi-specialized). What is not certain is whether a scribe who writes tablets relating to different economic spheres is of higher status, because he oversees and logs various sectors of the Pylian economy,Footnote 101 or of lower status, since he is forced to serve any recording need presented, big or small.Footnote 102 Seen differently, if, indeed, there were professional scribes in Pylos, we would expect them to coexist alongside the literate administrators.Footnote 103 This, again, would unavoidably lead to a scribal output even greater than the one preserved. Due to the lack of space suitable for extensive archival activities, the potential of concentrating tablets in such a small area for a few days or weeks, would immediately present storage and filing problems.
However, the prevailing theory is that most of the scribes did not belong to a professional class, but rather that they were literate officials. One of the main arguments is the tablets per Hand ratio, observed for many scribes.Footnote 104 At Pylos, for example, Hands 3/603, 4/604, 6/606, 11/611, 12/612, 13/613, 22/622, 24/624, 25/625, 31/631, 32/632, 33/633, 34/634, 42/642, and 45/645 have written fewer than twenty tablets each. Several of them are responsible for only two or three tablets, a very low production rate for professional scribes. In the latter scenario, it is safe to presume that the task and the governing limitations would turn the maintenance and updating of a central archive into a hard-to-follow process by the local officials. We are inclined to assume that their administrative duties would precede their supplementary role as literate state employees, regardless of the implications this situation would have presented for the management of an archival unit.
Conclusions
By re-evaluating textual and archaeological evidence we are one step closer to questioning the dominant view that Rooms 7 and 8 were a storage area for tablets, which were kept here for a relatively long period, from several months or up to a year. Beginning with the material of the tablets, clay is certainly one of the most logical choices for their manufacture, since it is available in large quantities and there are specialist craftsmen or less specialised workers able to process the raw material in order to produce writing documents (tablets, sealings, labels), not just in normal circumstances but also in the face of extreme and non-predictable events, in times of mass recordings of products and transactions, or in relation to administrative practices demanding constant scribal registration of personnel, materials, and services. The clay fabric of the Pylos tablets shows significant differentiations, implying variations in the materials used, their place of origin, or the production mode. The sizes and forms of the Linear B tablets also present considerable variations, as do the fabrication techniques. The lack of interest in a standardised way of making either their fabric or their shape is not something we would expect in an organised recording procedure, especially if it was aimed at the renewal or creation of a stable archive. The development of certain characteristics in the fabrication, shape, weight, and form would have been a logical choice, which is not observed in this archival collection. Complementary to the type of tablets used, the ratio of palm-leaf to page-shaped tablets represented in the AC is of equal importance. Only one-fifth of all written documents in Room 8 and roughly one-tenth of all preserved tablets in Room 7 are page-shaped tablets. Given that page-shaped tablets are used mostly for summaries and compilations, we would expect a higher percentage of collective documents of this kind inside the so-called Archives Complex. The concentration of a large number of documents in this area reveals intense scribal activities, but it does not presuppose its constant use for such a purpose or the duration of the recordings made in situ.
Noteworthy are textual peculiarities, such as differentiation in syntax and morphology, which would not facilitate the need for ordering and organising the output of tablets. If we add the possible indications that clay documents were checked, reread, and reused by more than one administrator, we are left with serious doubts as to how widely their context and thematic arrangement were understood by third parties. The formation and the inscription of the tablets, sealings, and labels were mainly the result of a practical spirit. The simplistic schemes and the minimal requirements in terminology used by the scribes remind us of improvisations or acts of necessity, so hinting at the temporary nature of these bureaucratic documents. The slipshod work produced by the scribes suggests a hasty execution of the tablets. It was not a slow, careful processing of transactions, contributions, and obligations. Instead, the speedy scribal procedure corroborates the casual treatment of the texts by the literate officials.
Overall, scribal conventions may be embedded in the text layout on both palm-leaf and page-shaped tablets. Nonetheless, the fact remains that there is no obvious consistency in their use that would allow us to connect them to well-defined, systematic morphological, and syntactical patterns. This intentional tolerance of the attested scribal idiosyncrasies is not easy to justify when it comes to the creation of a centralised assemblage of clay documents which needs to be thoroughly understood primarily by the master scribe and secondarily by any literate administrator of the palace. Clearly, the absence of common morphological and syntactical traits, resulting from the lack of official scribal rules, is not something to be expected from a centralised assemblage of documents.
While not underestimating the quantity of tablets, sealings, and labels found to date in this architectural unit, we are obliged to consider the limited size of the building complex and the information presented so far about the scribal production of the palace functionaries. All Linear B finds point to different sets of documents, organised on the basis of indiscernible criteria, to satisfy the needs of the officials working in Rooms 7 and 8. The fragmented condition and the dispersal of many tablets in Room 8 prevent us from identifying clear reasons for these grouping patterns. Since we know so little about the administrative approaches which defined the filing arrangement in the AC, we are also left wondering whether an archivist or master scribe was ever in need of such a conscious and steady effort to deal with this written production at all. Considering the tablet output of the total number of twenty–twenty-three scribes, who were dispatching their records to the AC, and the limited capacity of this building, there is reasonable doubt as to whether this volume of Linear B documents remained intact for a long period and, if so, whether their content was considered valuable enough to be organised in a systematic, scrutinised way.
There is also no clear evidence for literate administrators with permanent duties, mainly because the archaeological data refer to the period just before the destruction of the palace. Many of the surviving texts appear to have been inscribed by a relatively small number of literate individuals. Still, the evidence for a class of professional scribes remains scarce and dubitable. Hence, the organisation of an archival unit would not have been easy without the appropriate functionaries. On the other hand, the use of literate administrators as scribes would pose various problems. The task and the limitations would turn the maintenance and updating of a central archive into a hard-to-follow process by the local officials. We are inclined to assume that their administrative duties would precede their supplementary role as literate state employees, regardless of the implications of this situation for the management of an archival unit.
The deposits of Linear B tablets at different points in Rooms 7 and 8 could be easily justified by viewing these spaces as an architectural and working unit for the in situ production of clay documents, including places for literate officials to perform recording activities, as well as maintenance/support areas for the storage and sun/air drying of the newly made tablets. Of course, these practices did not necessarily relate to archival procedures in a broad sense. They probably represent a general need for controlling or keeping up with various stages of the transactions and distributions recorded, up to the point of their completion. It seems logical to think that the scribal output and writing habits of the Pylian officials and scribes were mainly intended to record the course of the transactions and economic activities, and not to preserve their memory, at least not for a long period after their execution. Consequently, instead of regarding Rooms 7 and 8 at Pylos as an archive, it seems more reasonable to view them as a temporary storeroom or a short-term assemblage point for the officials working in situ and their subsequent scribal production. Several economic activities, such as an inventory, a harvest, a seasonal or alternate system of labour (e.g. ta-ra-si-ja), a standardised tax payment period, even a sudden event or a state of emergency,Footnote 105 could prompt scribal production of the kind observed in the archaeological finds from the AC. Ceremonial/ritual activities, regular or impromptu, might likewise necessitate the use of written documents, especially when feasting events were involved. During such procedures, Linear B documents would serve as a means of communication, temporary aides-mémoire, and direct evidence for the supervision of the necessary transactions and personnel, as we would expect for tablet assemblages. Rooms 7 and 8 would form an architectural complex suitable for the coordination of these administrative efforts. Nevertheless, it is now debatable whether these rooms could achieve the goal of maintaining a huge quantity of written documents, for a period of more than a few days or weeks.
