If you click on the link Textual search in the menu on the left, a page will open with prompts for the tools to conduct an advanced search for a specific word(s) or portion(s) of text in the inscriptions.
In the text box you can enter the text you are searching for by typing one or more strings of characters:
- Search for a word: write the word in the text box; choose the option Showing results as: word list; click on search.
- Search for textual portions: write the words in sequence with a space between each word; choose Showing results as: text portions; choose a value from the Word range dropdown menu (see below); then click on search. This search will retrieve the list of textual portions searched for, according to the range parameter you chose.
- Search for two or more words: this search function operates in the same way as (2) except that you must choose the option Showing results as: word list. This search will retrieve the results of the words searched for, displayed in separate alphabetical lists by their initial letter.
Clitic and special characters
The clitic in the text is equivalent to a space, so if you type b ʾlmqh
you will find the occurrences of the textual portion b ʾlmqh
but also of b-ʾlmqh
. Therefore, the hyphen sign must not be typed in the box.
Special characters – in particular letters with diacritic marks – can be inserted by means of the virtual panel that is displayed by clicking on the icon below the text box.
(1.), (2.) and (3.) show how you can search for words when the strings of characters that you have entered in the text box coincide with words. However, you can also search for all the words that begin with, end with, begin and end with, or contain a string of characters. To do this, we have provided a function using the wildcard *. Note that the search finds also words beginning, ending or containing lacunae (see also Word lists
- Search for a word that begins with a specific string of characters: type the initial string followed by *. For example, if you enter ʾlm* the search results will include "ʾlm[... ...]", "ʾlm", "ʾlmnbṭ" etc.
- Search for a word that ends with a specific string of characters: type the string of characters preceded by *. For example, if you type *qh the results will include "[... ...]mqh", "ʾlmqh", "ḏmrwqh" etc.
- Search for a word that begins and ends with specific strings of characters: type a * preceded and followed by strings of characters. For example, if you type m*m the results will include "m[…]m", "mʾhm", "mʾs²mrm", etc.
- Search for a word that contains a specific string of characters: type a string of characters preceded and followed by *. For example, if you type *qh* the results will include "[... ...]wqhʾl", "ʾlmqh", "fqhy", "mwqhm", etc.
Note that the wildcard can be used also in the search for textual portions, combining the different cases; for example: s¹b* w *n
The isolated wildcard (alone: *
; or between spaces: byt * ywm
) is equivalent to a word. Therefore, note that if you search for the wildcard * alone, you will retrieve all the words; if you apply some restrictions (see Filters), you will retrieve all the words that satisfy the criteria selected.
The search for words or portions of text can be refined using one or more filters that will be applied to the meta-information encoded in the text of the inscriptions (Onomastic categories
and Textual portions
) or relative to the description of different features of the epigraphs (Language and script
, Textual typology
). For a more detailed description of the filters, see Editorial criteria
The different types of filters are grouped into expandable sections (Onomastic categories
, Textual portions
, Language and Script
, etc.); they follow the AND logic. For example, if you search for the string *nb*
and select the filters male
in the section Onomastic categories
in the Language and Script section
, the search will retrieve only the words that are male names containing the pattern nb
in inscriptions in Hadramitic language. The sections, except for Onomastic categories
, contain one or more menus from which you can select a filter. Also the menus of the same section follow the AND logic. For example, if you search for the pattern *nb*
and – in the Language and Script
section – select the filters Language-->Hadramitic
and Writing technique-->Incision
, the search will list only the words containing the pattern *nb*
that occur inscriptions written in the Hadramitic language and incised. So taking into account the AND logic between the sections and the menus, in case you were searching for the pattern *nb*
with filter male
in the section Onomastic categories
and the filters Language-->Hadramitic
and Writing technique-->Incision
in the section Languange and Script
, the search would retrieve only the masculine names containing *nb*
, found in inscriptions in Hadramitic language and incised.
In the section Onomastic categories
, the filters are check boxes, and you can select one or more filters at the same time which are in a AND and/or OR relation: when selected horizontally, the filters follow the AND logic, when selected vertically, they follow the OR logic. Examples:
- If you select the Gender filter, the system will retrieve the words which are a personal name, irrespective of whether they are masculine of feminine, royal names or not, patronymics/ matronymics or not.
- If you select the Male and Female filter, the system will retrieve the words which are a masculine or feminine personal names, irrespective of whether they are royal names or not and whether they are patronymics or not.
- If you select the Male and the Royal filter, the system will retrieve the words which are a masculine personal name, and at the same times are names of kings or of members of the royal family irrespective of whether they are patronymics or not.
- If you select the Male filter and the not Royal filter, the system will retrieve the words which are a masculine personal name, but will exclude the names of kings and of the members of the royal family, irrespective of whether they are patronymics or not.
- If you select the Male filter, the not Royal filter and the e not patronymic/matronymic filter, the system will retrieve the words which are a masculine personal name, but will exclude the names of kings and of the members of the royal family and the patronymics.
- If you select the Patronymic/Matronymic filter, the Lineage name filter and the Theonym filter, the system will list the words which are names of kings, queens and members of royal family, plus the words which are names of lineage, plus the words which are names of gods, in a single list.
And so on.
You should note the function Filter by
, which allows you to narrow your search to onomastics only
or lexicon only
. If you choose this option you cannot use the filters of the Onomastic categories
, because you have already chosen to restrict your search to words or portions of text that are names (independently of their type: Onomastics only
) or are not names (Lexicon only
). The other filters can still be applied.
The word range
is the maximum number of words that can intervene between the first word and the last word you are searching for (not counting the first word). Note that to find consistent results, the range number must be the same or higher than the number of spaces in the textual portion searched for. Some examples:
- if you do not select any range, the words are found in textual portions composed solely by the word you searched for, in the exact order they were typed; the same happens if you select the equivalent range to the number of spaces between the typed words. I.e:
- if you search for byt ʾlmqh ywm and you choose Word range: 2, the search will find solely the very textual portion “byt ʾlmqh ywm”.
- if you search for byt ʾlmqh ywm and you choose Word range: 4, the search will find textual portions composed of 3, 4, or 5 words of which byt is the first one and ywm is the last one: "byt ʾlmqh ywm"; "byt ʾlmqh b-ywm"; "byt ʾlmqh w-b ywm".
Order of the strings of characters
The order in which the strings of characters (in the case of two or more) is written is very important. The occurrences that will be identified will have as their first word the first string of characters, followed by the others. For example, if you search for gbʾw w
(either as a word list
or as a text portion
) with the range 1, the results will include gbʾw w
but not w gbʾw
BE CAREFUL: multiple tabs/windows open in the same browser
- If you open a new tab or a new window in the same browser by clicking on the Textual search button or on one of the Indexes from the left menu bar, and perform a search and/or apply filters, the search parameters and the filters will not influence the new search.
- If you open more than one tab or window in the same browser by clicking on the breadcrumbs and you use the Search tool or consult the Indices applying filters, the browser will apply to all the tabs or windows the filters you have activated by means the last command, in whichever tab or window.
For example, if in tab A you filter the textual search by Laguage
In case 1), when you open a tab B and perform a new search with Language
parameter, this parameter will not be applied to the search in the tab A;
In case 2), if you open a tab B where you filter a new textual search by Language
or any other filter, when you come back to the tab A, the results will be those provided by the last filters applied, to say those you selected in tab B.
The search function will generate a list of results identifying the inscriptions that contain the words or text portions of interest. The results are presented in a concise and legible manner that allows you to see the distribution of the words at a glance, to view the contexts in which they appear and then to access the inscription in few steps.
- A list of the words or portions of text that meet the search criteria is generated, grouped according to this schema: word, language, indication if the word is a noun or a name of a particular onomastic category, number of occurrences. E.g.: search for rym*:
|Epithet of god, first position in compound name||1|
|Epithet of god||123|
- By clicking on the link provided by the number of occurrences of a given word or portion of text, you can pass to a list of these occurrences. In this list, the specific language of the inscription, its siglum in alphabetical order and the context within which the specific word or textual portion appears in each inscription are provided, according to this schema:
|Central Middle Sabaic||CIAS 32.81/b 9||[... ... w-b Tʾlb] Rymm w-mnḍḥ-hmw bʿl|
|Central Middle Sabaic||CIAS 34.51/o 6||hqny s²ym-hmw Tʾ4lb Rymm bʿl 5Ḥdṯnn ḏn ṣ6lmn|
Thumbnails images of the epigraphs can be viewed by clicking on activate image preview
just above the list of occurrences.
Once your search has generated a list of occurrences, by clicking on the siglum of the inscription, you can retrieve the catalogue entry for that inscription, in which you will see the word/textual portion highlighted in bold.
Order of the result list
The search results will be listed according to this alphabetical order, adopted throughout the archive:
a, ʾ, ʿ, b, c, d, ḏ, ḍ, e, f, g, ġ, h, ḫ, ḥ, I, j, k, l, m, n, o, p, q, r, s, s¹, s², s³, ṣ, t, ṯ, ṭ, u, v, w, x, y, z, ẓ.
The numerical order is: 1, 10, 11, 110, 111, 2, 20, 240, 3, 300, 33 and so on.
Difference between the result lists generated by the word list option and the result lists generated by the text portion option
The difference between a search using the word list and a search using text portions will emerge clearly from the lists that are generated (see Search Modalities).
When you conduct a search by word list
, all of the words corresponding to the individual strings of characters will be searched for within the limits of the word range you have selected. For example, if you are looking for the two strings s¹b* w
with range 2, the list of results will include (among others) the following words (note: the DASI database is being constantly updated and therefore the number of occurrences will be accurate for the moment that this search was conducted
- "s¹bʾt" with only 3 occurrences in Sabaic, whereas if you were searching for the string s¹b* alone the list of results would include 11 occurrences
- "s¹bʾw" with only 9 occurrences in Sabaic rather than the 18 that would be retrieved if you were looking for the string s¹b* alone
- "w" with only 292 occurrences in Sabaic, instead of the 7,320 if you were looking for the string w alone.
And so on.
When you are conducting a search using text portions
, the entire portions of text that contain all of the strings specified in your search at the range defined in the search criteria will be listed. For example, if you are looking for the two strings s¹b* w
at a range of 2, the list of results will include the following portions of text (the number of occurrences will be accurate for the time that the search was conducted):
- "s¹bʾt w" with 3 occurrences in Sabaic
- "s¹bʾw w" with 9 occurrences in Sabaic
And so on.
It may be noted that the number of occurrences of a portion of text will be equal to the number of occurrences identified in a word list
search of the first string of characters.