Constituent Labels

June 2010 - G. Buccellati
[August 2022 - M. De Pietri]
A Digital Monograph
The home for this page is GRAMMAR

4.2 Ranks
4.3 Rank 1: Project
4.4 Rank 2: Constituent prefix (book)
     (1) First Segment.
     (2) Second Segment.
4.5 Rank 3: Constituent proper
4.6 Rank 4: Component
4.7 Rank 5: Sub-component
4.8 Constituent label: Summary
4.9 Incidentals

4.1 Labels

     A label is the identifier of a constituent. It occurs invariably as an alpha-numeric code which is derived from a sequential log. It identifies uniquely any given constituent, in the sense that it remains forever linked with the pertinent constituent as defined at the moment of excavation; there is no other overall and required registration number assigned sub­se­quently to any constituent.
     This label is always a generic label: it attributes the minimum possible level of specificity to an element, such as feature or item. Whenever the term “label” is used by itself, it refers to such a generic label.
     Other labels are available for use, but they are not required, and they never replace the generic label. In particular, a specific label attributes to a label the specificity inherent in a definition, adding an indexing factor that establishes a sequential list. A nested progression of specificity is possible, e.g., ceramic vessel, jar, oil jar, etc. Specific labels are derived from a lexicon of variants, and are properly attributed to elements only, since referents do not have higher levels of specificity. They will be discussed below, ....

4.2 Ranks

     A full label consists of a maximum of five segments, which I call “ranks,” as follows:
          Rank 1: Project and archive
          Rank 2: Book
          Rank 3: Constituent
          Rank 4: Component
          Rank 5: Sub-component
     The sequence of these ranks is fixed, and Ranks 2 and 3 are obligatory, while the others are optional. Note that labels are case sensitive.

Back to top

4.3 Rank 1: Project

     A full label consists of a maximum of five segments, which I call “ranks,” as follows:
     The first rank is optional, and refers to the Project. Typically, this refers to an excavation site, cited in an abbreviated form, e.g., MZ for Mozan. It may also stand for a project that does not involve excavation, such as a survey, and in that case an analogous kind of abbreviation will be chosen. Rank 1 needs to be included in the label only where the constituent in question may become separated from the site or its archive: thus objects, slides or prints should have the site as part of the label, but not a benchmark.

Back to top

4.4 Rank 2: Constituent prefix (book)

     Rank 2 is obligatory. It corresponds to an excavation unit or an area, and therefore also to a book (see below, 15.2).
     It consists of an alpha-numeric code which includes in turn two segments. Note that in this rank letters are always in upper case.

Back to top

(1) First Segment.

     The first segment of the code for Rank 1 refers to either a topographic zone (for a list of areas in Mozan1 see Fig. 4-1) or a type of analysis.
     The letters between A and S are reserved for topographic zones, and are the only ones that are used to define constituents proper. In our system, the letter I is omitted because it is graphically confusing, and M is omitted because it may be used for “Mozan.” The letter O refers to the Outer City at Mozan/ Urkesh. (This particular feature may not pertain to other sites. However, it is in the logic of the system to reserve the area prefix O for Outer City, and to restrict the use of topographic zones to the prefixes A through S.) So every topographic zone in the Outer City includes always an initial O (this is the only instance where the code for zone contains two letters); zones in the Outer City approximately mirror the arrangement of zones on the High Mound. The letters S and OS are reserved for Special operations at any place on the High Mound or Outer City respectively.
     It is a general practice to label operations either sequentially in the chronological order in which they are begun, or according to the grid. The sequence by topographical features as we have done in Mozan seems preferable, because it provides for every single item or feature an immediate sense of orientation within the site as a whole, while avoiding the complexities of labels based on the geometrical grid.
     The letters that follow in the alphabet (from T to Z), and other two letter combinations besides the O sequence, are used to define directories (see below, 15.2), but not constituents, except for the following:

Z1 sitewide surface items
Z2 excavated items without number (label has been lost, but items, generally only objects, are typologically significant)
Z10 wall conservation project – introduction
Z11 wall conservation project – Palace – used with referents only
Z12 wall conservation project – Temple Terrace – used with referents only
Z13 wall conservation project – Temple – used with referents only
Z30 site presentation – general and itinerary – used with referents only
Z31 site presentation – signposts
Z32 site presentation – panoramas
Z33 site presentation – reading stands

Fig. 4-1 Topographic Zones at Tell Mozan

     Fig. 4-1. Zones of High Mound and Outer City
The letters may be chosen sequentially, as with temple BA at Mozan (where this particular structure, the temple, was the first excavated in this area), or with reference to the functional nature of the particular building, as with building AK also at Mozan (where K stand for the Sumerian word kishib “seal,” with reference to the high number of seal impressions found in this building).

Back to top

(2) Second Segment.

     The second segment of the code for Rank 2 refers either to a subdivision of a topo­graphic zone, i.e., an area, or to a subtype of analysis. For an area within a topographic zone, the code is numeric (from 1 to 99, see chapter 15.3) if it refers to an excavation unit seen purely as volumetric area, i.e., an area which is defined exclusively with reference to boundaries that match preset loci, e.g. A1. It is instead alphabetic if it refers to an area that has an architectural or functional integrity, e.g., AK.
     Given the reliance on absolute coordinates for volumetric control, rather than on the grid conceived as a physical overlay in the ground, excavations units need not have predefined geometric shapes, and are generally quite asymmetrical. In fact, they may also overlap. Fig. 4-2 represents the situation as it occurs in Area AK at Mozan. Label AK refers to the overall area as defined functionally in terms of the architectural layout of the Royal Building. The other labels (A1, etc.) refer to purely volumetric entities defined operationally in terms of logistic needs and considerations. Thus A8 overlaps with A6 and A10 which were expanded at a given point to subsume what had originated as an independent operation.

AK units

Fig. 4-2. Excavation units within AK

Back to top

4.5 Rank 3: Constituent proper

     Rank 3 is obligatory. It consists of the constituent proper, which is identified by a prefix (a single lower case letter corresponding to the type of element or referent) and a label index (a number referring to the sequential order of constituents). The maximum range allowed for the sequential number is 4 digits, so that the highest number allowed is 9999.
     With items, the identifying letter i is regularly replaced by a period in the label;1 it is retained in other instances, e.g., when only a partial label is used (thus, in a context devoted to A1 one will write i12 instead of A1.12), and as directory names.
     Strata and phases have a distinctive labeling system. Within any given book, the strata/phases are generally taken from a different book (e.g., within J3, the strata/phases are taken from JP). It is also necessary to indicate the generation to which the strata/phase sequence belongs. Accordingly, the reference book (JP in our example) and the generation (indicated with a capital letter) must be given as a label index preceded by a hyphen. Thus, J3s1-JPA indicates that the current strata sequence in book J3 is generation A of book JP.
     Note that the numeric component of templates (t) is always identical to the numeric component of the equivalent view (v).

Back to top

4.6 Rank 4: Component

     Rank 4 is optional in some cases, obligatory in others. It consists of an alpha-numeric label, which includes in turn two segments, of which the second is optional. Note that, in this rank, letters are always in lower case. Components are found only for elements, i.e., for features, items and q-lots.
     The first segment is the typological component (see above, 12.3 and 13.2). It consists of a single lower case letter, which refers to one of three possible types of typological components, as follows:
     -b bones
     -i items other than bones and pottery (replaced by a dot)
     -p pottery
     Thus A1q2-p refers to all (unsorted) sherds of lot A1q2. The same would apply to a feature, e.g., A1f2.1 might refer to a bead found within a wall that is being dismantled. Should there be more than one bead, each would be identified by a sequential number, such as A1f2.1.3 for the third bead.
     The second segment is a number referring to an individual component, and the number is taken from a sequential order from 1 to 99. Thus A1q2.3 refers to the third item of lot A1q2, and A1q2-b3 to the third bone of lot A1q2.
     The series of numbers from 70 to 99 is used for classes of components that are analyzed only as groups, e.g., body sherds in a pottery lot. Thus A1q2-p70 refers to a group of body sherds of a given type (e.g., “Early Trans-Caucasian”) from lot A1q2.
     The series of numbers from 9000-9999 is used for classes of elements that present problems discovered after the standard label has been assigned, and cannot be corrected on the original. This applies, for example, to old drawings that do not have a component number. Thus the drawing of sherd A1q2-p (no sherd number after p) would receive a label A1q9000-p1, and other unnumbered sherds in the same lot would be sequentially numbered in the same way; q9000 would correspond to q2, meaning that sherds from other lots would receive a different lot number. A note would have to be entered under both q2 and q9000.
     The number 0 is reserved for a stratum which is mixed or unclear as to statigraphy, as, for example, with the excavation of a purely "volumetric" entity such as a baulk. The number 9s is reserved for teh phase of that same stratum.
     The component is separated from the element by a hyphen. As for constituents proper, the code –i is replaced by a period.

     Note that the period used instead of the letter i makes the label easier to read. It is also an apt abbreviation in that an item is in fact a point triangulated individually, i. e., with unique coordinates.

Back to top

4.7 Rank 5: Sub-component

     Rank 5 is optional. It consists of a single numeric code, from 1 to 99, that serves as the quantitative identifier – i.e., it gives the number of an individual component found within either an element or a component. Thus A1q2.3.5 refers to the fifth subcomponent of item 3 within lot A1q2. For instance, if a group of beads has been collected as a lot (A1q2.3), when the time to describe the individual beads comes, each will be assigned a sequential number, so that A1q2.3.5 would refer to the fifth bead within that lot. Another example is the numeration of individual sherds comprising a vessel or a larger sherd.
     If the quantitative subcomponent is added to an element, then it is preceded by two dots. Thus A1.2..4 refers to the fourth component of item A1.2.

Back to top

4.8 Constituent label: Summary

     Fig. 4-3 provides a synoptic chart of all possible combinations that may be found to identify a minimal constituent.

Fig. 4-3 Synopsis of constituent labels

  Ranks Labels
  2 3 4 5
  Book Prefix Label Index Component/Suffix Sub-component/Qualification
item A1 i 2 (written .2)     A1.2
A1 i 2 (written .2) -i3 (written .3)   A1.2.3
A1 i 2 (written .2) -i3 (written .3) .4 A1.2.3.4
A1 i 2 (written .2)     A1.2.4
template A1 t 123 a   A1t123a
view A1 v 123 a   A1v123a
q-lot A1 q 2     A1q2
q-item A1 q 2 -i3 (written .3)   A1q2.3
A1 q 2 -i3 (written .3) .4 A1q.3.4
A1 q 2 -b3   A1q2-b3
A1 q 2 -b3 .4 A1q3-b3.4
A1 q 2 -p3   A1q2-p3
A1 q 2 -p3 .4 A1q2-p3.4
graphic constit. V19 d 24 13   V19d2413
W19 c 12 01   W19c1201
strata and phases J3 s 11   -JPA J3s11-JPA
J3 s 11 a -JPA J3s11a-JPA
J3 h 2 a -JPA J3h2a-JPA
special numeric sequences A1 q 2 -p70(-99) body sherds A1q2-p70
A1 q 9000(-9999) problems with q-lot identification A1q9000-p1
A1 s 0 unclear or mixed strata J3s0-JPD
A1 h 9s unclear or mixed phase J3h9s-JPD

     Dark shaded cells represent obligatory fields, light shaded cells represent fields that are obligatory only outside the pertinent book

     The obligatory ranks are 2 and 3, so that minimal examples of legal constituent labels are as follows: A1f1, A1.10, AKr1001, OS12.101.
     With the optional addition of rank 1, where needed (e.g., when the label is written on an object), we would have: MZ A1.10.
     With the optional addition of ranks 4 and/or 5, we would have: MZ A10q8.1, MZ A10q8p1, A10q8p1-3, OBq987.66.1.
     For the constituent prefix of drawings and photographs see below, chapter 23.

Back to top

4.9 Incidentals

     As indicated, incidentals approximate the status of constituents, even though they are structurally quite distinct. I have called them para-constituents (12.2) as a way to refer both to the similarities and the differences between them and constituents proper. They are not, accordingly, part of a closed system, and thus no structural listing is possible as the one given above for constituents. An important difference is that there is no sequential numbering of incidentals. As a result, the label for incidentals is limited to the type, expressed by an alphabetic code preceded by a hyphen. For instance, -sg refers to “strategy,” and A1-sg refers to a discussion of strategy within A1.
     The codes used to denote incidentals belong, as I have stressed (see above, ...) to an open system, and only a lexical listing is possible. A list of codes currently in use at Mozan will be found in 17.4 (4), but any addition is possible to this list. The only formal constraint is that the code begin with a hyphen and contain no space.

Back to top