• Inscription #

We have tried to offer documentation and guidance for the use of the Inscriptions of Roman Tripolitania 2009 online corpus. If you have any difficulty that is not covered below, please report the problem to our contact address.


Category search help

Select one or more options on which to search:



Use the 'radio buttons' on the left to choose between the two options: either enter a 'to' and 'from' year in the two boxes provided—use negative numbers (preceded by a minus-sign or hyphen) for B.C. years— or pick a century from the drop-down list (recommended).


Object Type / Text Type

Select an item from the drop-down menu. Note that some categories are relatively uncommon, so may not return any hits in combination with other options.


Dating Criteria

Select an item from the drop-down menu. Note as above that some criteria are relatively uncommon (the bulk of texts are dated by lettering, for example), so this option is probably best combined with date rather than the other options in this search page.


The thirteen chapters (and six sub-sections) that make up the sites in Tripolitania are listed in the drop-down menu. A more convenient way to browse this category might be via the Table of Contents by Location.



This is a simple text search, the only options being between Greek, Latin, and English. By default the text entered will be matched to the beginning of a word in the source files. For more detailed search options, use the advanced text search facility.


Text search help

The text search input page offers the option to search in Latin, Greek or English, and contains three input fields. Only one word should be entered into each field: the search facility does not currently support phrase searching.


Unicode Input (Greek)

To enter Greek text in Unicode you will need a polytonic Greek keyboard driver for your computer. (See the Digital Classicist wiki for advice on keyboard drivers for Mac/PC.) If you normally have a setting for Modern Greek on your computer you can also use this for unaccented Greek, but you will not be able to enter the polytonic accents and breathings. If you do not have a Greek keyboard setting, it may be easier to type using Beta Code.


Beta Code Input (Greek)

Beta Code is a simple system for encoding Greek using 24 characters of the Latin alphabet to represent the Greek letters. It is principally used by the TLG, PHI, Perseus, and Duke Databank projects. The character equivalents in the Beta Code alphabet are:

α β γ δ ε ζ η θ ι κ λ μ ν ξ ο π ρ σ/ς τ υ φ χ ψ ω

Beta Code may be typed into the search form in upper- or lower-case. Diacritic-sentitive searching is not possible in Beta Code entry mode.


Regularised lemmata

If this option is checked, your search will be on the dictionary forms of words (as in the word index) and the nominatives of names (as in the name index) rather than on the oblique and possibly irregular forms found in the texts themselves. Not all words have been unambiguously lemmatised, so this mode will necessarily be more limited than the normal search mode.


Whole word

If this option is checked, your search term will be matched only if it is the whole word as it appears in the text (or lemma). For example, a normal search for "και" will include all results that begin with those three letters, such as καινός and Καῖσαρ; if "Whole word" is checked, the same search will only find the conjunctive καί.


Words without supplement

If this option is checked, your search term will only be matched if the characters actually appear on the stone. That is to say, words including editorial supplements (due to damage, error, abbreviation, etc.) will not be included in the search results.


Interpreted text only

If this option is checked, only characters that form part of identifiable words will be included in the search; that is to say all characters except those which the editor has marked as unresolved (in upper-case in the standard edition).


With diacritics

When this option is selected your search term will be treated as diacritics-sensitive. This means that if you search for "ἀγ", for instance, the search will return ἀγαθός, ἀγλαός, and ἀγωνοθέτης, but not ἄγαλμα or ἄγγελος (accent on the alpha), or ἁγνός (different breathing mark). An unaccented letter in your search term will only match an unaccented letter in the text.


In names only

If this option is checked, your search will only return results in text that has been identified as a personal name (whether of a real person, a mythical figure, or a divine entity). Words, placenames, etc. will not be matched.


Greek font help

In order to view the Greek, and particularly some of the epigraphic sigla, in this publication you will need a Unicode-compatible operating system and web browser, and at least one font with good polytonic Greek Unicode coverage. For more information on Unicode and Greek fonts, please see one of the following pages:


Links help

There are a variety of different links in the IRT website. In most cases hovering over the link will cause a tool-tip to appear (depending on your browser) giving more information on the destination of the link. A few particular kinds of links are described below.



The thumbnail images on this site (small images, principally photographs or sketches of inscriptions below the editions themselves, re-sized to be 100 pixels along their longest dimension), act as selectable links to the full-size images they represent. Selecting this link will open the full-size image (1000 pixels along the longest dimension) in a new window sized to the image.


Jump to inscription box

Enter the number of the inscription (one to three digits plus optional letter, e.g. 4 or 939a) into this box, and either hit return or select the button labelled "Go to inscription:". The three-digit number that makes up the filename of the inscription is derived automatically from numbers below 100, so need not be typed in full into the box.


Contextual index links

In the word and name indices, the headwords on the left hand side are links to a contextual index (also known in linguistic terms as a concordance). Selecting any of these words or names will take you to a page in which all instances of the word in question are shown in their immediately surrounding context (usually half a dozen words to the left and right). You may also link directly to the inscriptions in which the word or name appears.


Downloadable XML

At the bottom of each inscription page in this corpus is a link allowing users to download the inscription in EpiDoc XML format. There is also a link to an online version of the EpiDoc DTD (version 6). Selecting any of these links in your browser in the normal way may cause the raw XML to appear on the screen, or an error to occur, depending on the vagaries of your web browser. It is probably more useful to you to be able to download this file to your computer and open it with an XML editor that will format the contents for you, allow you to modify the file, validate against the DTD, etc. To download the file, right-click (or ctrl-click on a Macintosh) on the link and select "save link target as..." (vel sim.) from the menu that pops up. The browser will then ask you where on your local machine you want to save the file in question.

Note that you may also download the EpiDoc files from the XML Repository directly, see XML Repository page.

Epigraphic text display options

There are three options for the display of the epigraphic text field; select the buttons above the text to move between the views.

"Edition" view

The default view. In this view the text appears in epigraphic lines (or verse lines, in a few cases), in the form epigraphers would expect to see a printed edition: spaces between words, lower-case and accented characters, restorations and observations indicated using the usual Leiden brackets and sigla, etc. In most cases we do not use modern punctuation unless the text is particularly literary. Lines are numbered every 5.

"Diplomatic" view

In this view we present only the characters and signs as seen on the stone, interpreted only in as far as one letter-form is prefered over an other. All letters are in upper-case, there are no diacritics or spaces between words, and no characters are restored. Abbreviations and errors are neither marked nor resolved. In most cases, the diplomatic edition should look very similar to the preliminary transcription of the text in an epigrapher's notebook.

"EpiDoc (XML)" view

In the EpiDoc view we give a simplified rendering of how this would be marked up in EpiDoc XML. Elements and attributes have been coloured blue and orange for clarity. This is not the actual XML behind the edition (which you can download from the XML Repository or the link at the bottom of each screen), but only shows those tags that are required to express the distinctions traditional made by Leiden sigla. It is our expectation that this view will principally be of value as a teaching and learning tool, allowing the student to compare Leiden-expressed text with the EpiDoc XML equivalent.