Dependency Annotation Choices: Assessing Theoretical and Practical Issues of Universal Dependencies

Abstract

This article attempts to place dependency annotation options on a solid theoretical and applied footing. By verifying the validity of some basic choices of the current dependency reference framework, Universal Dependencies (UD), in a perspective of general annotation principles, we show how some choices can lead to inconsistencies and discontinuities, partly due to UD's alternation between syntax and semantics. For some constructions, we propose better suited alternative structures with a clear-cut distinction of syntax and semantics. We propose a classification of conception-oriented, annotatororiented, and finally, treebank end-useroriented considerations to be used in the creation of new annotation schemes.

The status of function words in dependency grammar: A critique of Universal Dependencies (UD)

“…Jill took a shower. As discussed in Gerdes and Kahane (2016), the UD decision to subordinate function words to content words predicts the UD approach to assume the a-analyses. However, UD actually advocates the b-analyses instead.…”

“…These goals can lead to contradictory conclusions, as shown by Gerdes and Kahane (2016) and as discussed at length below. 4 To refer to these goals efficiently, the designations in parentheses are used (e.g.…”

“…Gerdes and Kahane (2016) propose three finer-grained sets of annotation choices: conception-oriented considerations (including Desiderata 1 and 2 of the UD goals), annotator-oriented considerations (Desideratum 3 above), and end-user-oriented considerations (Desiderata 4, 5, and 6).…”

“…Gerdes and Kahane (2016) characterize this aspect of the UD annotation scheme as resulting in a catastrophe, as understood in the mathematical sense of catastrophe theory. A seemingly minor difference in the analyses results in drastic differences in the resulting structures.…”

SUD or Surface-Syntactic Universal Dependencies: An annotation scheme near-isomorphic to UD

“…The content-word-centric annotation is also problematic for the internal cohesion of the treebank (cf. the difficulty of coherently annotating complex prepositions that usually contain a content word, Gerdes & Kahane 2016) and it marks a break with syntactic traditions, where headedness is defined by distributional properties of individual words (Bloomfield 1933), see Section 2. 1 One of the central advantages of dependency grammar is the clear distinction of category (the POS, i.e.…”