Introduction to MultiTree
The trees in MultiTree are intended to be faithful representations of their sources. It can be difficult to capture a scholar's intentions in a graphical representation, but we have tried to add comments help to convey the difficulties and ambiguities the editor encountered when adding the tree. It is recommended that users try to refer to the original source for a better understanding of the scholar's analysis.
The presence of a tree does not constitute an endorsement of that tree. MultiTree simply aims to collect as many hypotheses about language relationships as possible so that users can compare them, and in many cases, see the progression of ideas from early hypotheses to newer ones.
Regarding the inclusion of contact languages (pidgins, creoles, and mixed languages) and language isolates
Although isolates have no known genetic affiliation, and the origins of contact languages are heavily contested, they have been included in the MultiTree database in order to make information about them available to scholars. "Trees" that include these languages do not reflect genetic affiliation unless this was the intention of the author.
Understanding the Trees
It is important to note that the nodes of a tree in MultiTree are not equivalent to the types, codes and names associated with them. Each tree is made up of nodes. A node may be of the type Stock, Subgroup, Language, Dialect or Dialect Group. In some cases, where a language has descendants, a node can be both a Subgroup and a Language. Each type of node has a different color.
A node's name is the name used by the scholar whose hypothesis the tree represents, and so the same entity may have different names from hypothesis to hypothesis. It will, however, have the same code in all of them, for this enables subgroups or languages whose names vary to be located. Language codes are ISO 639-3 codes, and thus standard. All other codes are Local Use codes. Subgroup codes are 4-letter codes. Dialect codes are the parent language's code plus three additional letters, separated by a dash. Both the subgroup and dialect codes were created specifically for MultiTree.
Minimum screen size:
Java (used in Star Tree):
Viewing trees requires at least version 1.5.0 of Java. Click here to download the
latest version of Java.
- Internet Explorer 6
- Firefox 1.5
- Opera 7
- Opera 9
- Firefox 1.5
- Opera 7.4
- Netscape 7
- Safari 1.4
This menu provides a starting place to explore MultiTree. It includes all the language families covered in MultiTree, organized by geographic region. Clicking on a family will give you a list of all the trees relating to that family. Because a number of families are found in more than one region, you may find the same families repeated.
Search using just one of these fields, or narrow the results using several of them. To switch between the Browse by Region view and the Search interface, click the word 'Search' next to the 'Browse' tab above the Browse by Region menu. The Search return button is in the blue bar at the bottom of the search form. You may use the Reset button to clear the form.
Search by name
Enter a language, subgroup or family name, or part of a name, here.
Search by code
Since scholars may use different names for the same subgroup or language, the Code search may yield more complete results. You can search the LINGUIST database for language codes to use in your search here. Alternatively, you can search the MultiTree database for a lect or subgroup and find its code in the search results. Then, you can perform another search in MultiTree using that code to ensure you return all relevant results.
If you are looking for a particular source, you can search by its title, or part of its title, here.
Search by a scholar's name, or a part of it.
Search for sources written in a certain language using ISO language codes. You can look up a language code here.
Some common language codes: English (eng), French (fra), German (deu)
Use the dropdown menu to select from the list of publication types, ranging from article to conference proceedings. Note: If the source was a single chapter of a book, select chapter.
Search by year, in the format YYYY. This can be useful for finding hypotheses from a certain time period. You can search for hypotheses from a range of time by entering a start and end year, or for a single year by entering that year in both fields.
The search results page lists all the matching hypotheses for your search term(s). The first line of each hypothesis in your results consists of the family name, author's last name and year of publication. Clicking on this title will open the tree in the StarTree viewer. The complete bibliographical reference is on the next line, followed by the total number of nodes in the tree, in green. Next to that total is a link to view the tree in xml, and a link to compare the tree. Clicking on 'compare' will select this tree as one of two to be opened together in the StarTree viewer, in the 'compare' mode. Before the StarTree viewer will open, a second tree/node must be selected using its 'compare' link. Trees which are selected to be compared will appear as light blue tabs at the top of the search screen, and can be unselected using the x next to their name in the tab.
Searching by Lect
If you searched for a language, subgroup or dialect name or code, you will see a list of every node in the tree which matches your search terms beneath the tree's bibliographical reference. Clicking on a node's name will open the tree, centered on that node. You can see each node's code, and whether it is considered a Subgroup, Language, Dialect or other type of node by the scholar. Selecting the details link will open a page with all the information on that code from the MultiTree database. The compare link works as above, except the tree will open centered on that specific node.
Tree Display: StarTree
The interface uses software called StarTree, which displays the tree in a hyperbolic viewer. The viewer allows the entire tree to fit in the window, but the parts at the edges of the window are smaller.
If you have trouble seeing the tree (nothing appears), you may need to update to a newer version of Java.
There are three ways of navigating through the tree.
- The explorer view on the left. This interface is linked to the Star Tree viewer, so as you open Subgroup folders and click on languages, the tree will focus on those nodes.
- The Star Tree viewer. Navigate by clicking and dragging areas you want to look at to the center of the viewer. You can also close and open subgroups by clicking on the little +/- in the lower right corner of each node. The full name of a node may not be visible unless you put your mouse over it or it is in the center, because the nodes shrink for better visibility of the whole tree.
- The Tree Internal Search. See below for further details.
Selecting the Options button in the upper right-hand corner of the StarTree window will allow you to customize the display of a tree. You may change the tree's orientation, so that the topmost node displays centered, to the left, right, etc. You may also: shrink or expand the distance between nodes, change the colors of the nodes and the background of the tree, change the font and its size, and select whether or not to display icons marking languages or dialects which are nearly extinct or extinct (according to the hypothesis's source). When comparing trees, you may change the display of an individual tree by selecting it. To return to the tree, simply click outside of the options window and it will collapse.
Searching Inside the Tree
To find a certain language or subgroup in the tree being viewed, select Names or Codes under the Tree Search heading on the left, then type in your search term. The field will show auto-completed matches as you type. If you don't see the Tree Search or if it disappears, drag the border at the bottom of the Tree Explorer upwards and it will become visible.
Node Info displays bibliographic information on a selected node, as well as the lect's name (including any alternates), code, type (e.g. Subgroup or Language), number of child languages, if any, and any comments made by the scholar about that lect. By selecting the thumbtack, you may save that node's Info and then open multiple Info tabs. You may also click on the green Details button, which will take you to a separate page displaying further information on the lect (see below).
The Language Information page, reached via the Details link, displays a list of all hypotheses in the MultiTree database which include the chosen lect. It also displays the lect's primary name, alternate names from all hypotheses, its code, the families in which it appears, any lects which appear as its parents and children, and any geographic information or comments available. The page also contains links to other sources of information including WALS, Google, Ethnologue, ODIN, and the LINGUIST List Database.
- A scholar login to enable scholars to edit their trees.
- A comment facility to discuss hypotheses.
To report a problem or to make a comment, email multitree(at)linguistlist.org