Figure 1.
The warp ikat process: making the resists.
Warp threads stretched on a frame, with partly tied ikat resists, on the island of Alor. Weavers use essentially the same method, using a similar frame, in all the areas considered in this study.
Figure 2.
The warp ikat process: weaving.
Sketch showing how a backstrap loom is used by weavers on Hainan Island to weave ikat cloth. Weavers use similar backstrap looms in all the areas considered in this study, though in most areas outside of Hainan weavers attach the warp beam to a fixed point rather than bracing it directly with the feet.
Figure 3.
Examples of Southeast Asian textiles with ikat decoration.
A: detail of Tai weft ikat cloth with bands of supplementary weft decoration, formerly part of a tubeskirt with a waist band above and a narrow band of plain cloth below, B: Meifu Li tubeskirt from Hainan Island, made up of 5 panels of warp ikat decorated cloth and one panel of supplementary weft decoration (lighter color), C: Iban tubeskirt from Borneo with warp ikat and supplementary warp decoration, single panel. D: Lamaholot tubeskirt from Ili Api (eastern Flores) made up of two panels of warp ikat and warp stripes, E: Timor tubeskirt from Belu regency, made up of 3 panels with warp ikat and warp stripes, F: tubeskirt made by Ngada people from central Flores, made up of three panels of warp ikat decorated fabric.
Figure 4.
Locations of warp ikat-weaving traditions (taxa).
Traditions included in this study are shown with solid red circles, warp-ikat traditions that were not included are shown as hollow circles.
Figure 5.
Examples of characters and their associated decorative features.
Figure 6.
Neighbornet plot of weaving taxa.
Clusters, which mostly correspond to geographic regions, are labeled at the edges.
Table 1.
Delta Scores for individual taxa, ordered from lowest to highest.
Figure 7.
Results of Bayesian analysis of weaving taxa.
70% majority rule cladogram of 1000 trees sampled from a 10,000,000-step MCMC search using flat priors, variable character-change rate. Clade frequencies are shown as percentages behind each node. Proposed taxonomy of weaving traditions shown by the bracketed taxa on the right hand side, based on clades with support of 90% or better.
Table 2.
Ancestral States for the mainland root weaving tradition (pSW) and the Austronesian root weaving tradition (pAW).
Figure 8.
Geographic distribution of warp ikat weaving clades and sub-clades.
Weaving clades correspond to those labeled in the Bayesian tree in Fig.7. The Daic and Austronesian weaving clades are indicated by dotted lines. The clades have some similarities (and some differences) with groupings identified in linguistic studies.
Figure 9.
Enlargement of the southeastern corner of Fig.8.
Weaving traditions with uncertain or blended affiliation are shown in grey.
Figure 10.
Reconstruction of the characters of the ancestral traditions of SEA warp ikat.
The ancestral characters of the pSW tradition, presumed to be located on the Asian mainland, and of the proto-Austronesian (pAW) tradition.
Figure 11.
Geometric motifs from cast bronze drums from the Dong-Son culture.
Figure 12.
Reconstruction of the ancestral states (motifs) of the Iban-Mindanao clade and the Benua-Austronesian clade.
The Iban and Mindanao weaving traditions form a sub-clade that has a rich group of 28 shared ancestral characters at its root, 25 of which are motifs (IM on Fig.7). In contrast the Iban-Mindanao and Benua traditions share a much smaller group of ancestral traditions at the root of the Benua-Austronesian clade (BA on Fig.7), despite the closer geographic proximity of the Iban and Benua weavers in Borneo.