Are embedding decisions independent?

Evidence from preposition(al) phrases

Abstract Full Paper (PDF)

One of the more difficult challenges in linguistics research concerns detecting how constraints might apply to the process of constructing phrases and clauses in natural language production. In previous work (Wallis 2019) we considered a number of operations modifying noun phrases, including sequential and embedded modification with postmodifying clauses. Notably, we found a pattern of a declining additive probability for each decision to embed postmodifying clauses, albeit a pattern that differed in speech and writing.

In this paper we use the same research paradigm to investigate the embedding of an altogether simpler structure: postmodifying nouns with prepositional phrases. These are approximately twice as frequent and structures exhibit as many as five levels of embedding in ICE-GB (two more than are found for clauses). Finally the embedding model is simplified because only one noun phrase can be found within each prepositional phrase. We discover different initial rates and patterns for common and proper nouns, and certain subsets of pronouns and numerals. Common nouns (80% of nouns in the corpus) do appear to generate a secular decline in the additive probability of embedded prepositional phrases, whereas the equivalent rate for proper nouns rises from a low initial probability, a fact that appears to be strongly affected by the presence of titles.

It may be generally assumed that like clauses, prepositional phrases are essentially independent units. However, we find evidence from a number of sources that indicate that some double-layered constructions may be being added as single units. In addition to titles, these constructions include schematic or idiomatic expressions whose head is an ‘indefinite’ pronoun or numeral.

How might parsing spoken data present greater challenges than parsing writing?

This is a very broad question, ultimately answered empirically by the performance of a particular parser.

However to predict performance, we might consider the types of structure that a parser is likely to find difficult and then examine a parsed corpus of speech and writing for key statistics.

Variables such as mean sentence length or main clause complexity are often cited as a proxy for parsing difficulty. However, sentence length and complexity are likely to be poor guides in this case. Spoken data is not split into sentences by the speaker, rather, utterance segmentation is a matter of transcriber/annotator choice. In order to improve performance, an annotator might simply increase the number of sentence subdivisions. Complexity ‘per sentence’ is similarly potentially misleading.

In the original London Lund Corpus (LLC), spoken data was split by speaker turns, and phonetic tone units were marked. In the case of speeches, speaker turns could be very long compound ‘run-on’ sentences. In practice, when texts were parsed, speaker turns might be split at coordinators or following a sentence adverbial.

In this discussion paper we will use the British Component of the International Corpus of English (ICE-GB, Nelson et al. 2002) as a test corpus of parsed speech and writing. It is worth noting that both components were parsed together by the same tools and research team.

A very clear difference between speech and writing in ICE-GB is to be found in the degree of self-correction. The mean rate of self-correction in ICE-GB spoken data is 3.5% of words (the rate for writing is 0.4%). The spoken genre with the lowest level of self-correction is broadcast news (0.7%). By contrast, student examination scripts have around 5% of words crossed out by writers, followed by social letters and student essays, which have around 0.8% of words marked for removal.

However, self-correction can be addressed at the annotation stage, by removing it from the input to the parser, parsing this simplified sentence, and reintegrating the output with the original corpus string. To identify issues of parsing complexity, therefore we need to consider the sentence minus any self-correction. Are there other factors that may make the input stream more difficult to parse than writing?

What might a corpus of parsed spoken data tell us about language?

Abstract Paper (PDF)

This paper summarises a methodological perspective towards corpus linguistics that is both unifying and critical. It emphasises that the processes involved in annotating corpora and carrying out research with corpora are fundamentally cyclic, i.e. involving both bottom-up and top-down processes. Knowledge is necessarily partial and refutable.

This perspective unifies ‘corpus-driven’ and ‘theory-driven’ research as two aspects of a research cycle. We identify three distinct but linked cyclical processes: annotation, abstraction and analysis. These cycles exist at different levels and perform distinct tasks, but are linked together such that the output of one feeds the input of the next.

This subdivision of research activity into integrated cycles is particularly important in the case of working with spoken data. The act of transcription is itself an annotation, and decisions to structurally identify distinct sentences are best understood as integral with parsing. Spoken data should be preferred in linguistic research, but current corpora are dominated by large amounts of written text. We point out that this is not a necessary aspect of corpus linguistics and introduce two parsed corpora containing spoken transcriptions.

We identify three types of evidence that can be obtained from a corpus: factual, frequency and interaction evidence, representing distinct logical statements about data. Each may exist at any level of the 3A hierarchy. Moreover, enriching the annotation of a corpus allows evidence to be drawn based on those richer annotations. We demonstrate this by discussing the parsing of a corpus of spoken language data and two recent pieces of research that illustrate this perspective.