The Extent of Repetition in Contract Language

Abstract

Contract language is repetitive (Anderson and Manns, 2017), but so is all language (Zipf, 1949). In this paper, we measure the extent to which contract language in English is repetitive compared with the language of other English language corpora. Contracts have much smaller vocabulary sizes compared with similarly sized non-contract corpora across multiple contract types, contain 1/5 th as many hapax legomena, pattern differently on a loglog plot, use fewer pronouns, and contain sentences that are about 20% more similar to one another than in other corpora. These suggest that the study of contracts in natural language processing controls for some linguistic phenomena and allows for more in depth study of others.

ConReader: Exploring Implicit Relations in Contracts for Contract Clause Extraction

“…Recently, studies grow increasing attention on CCE to extract clauses, which are complete units in contracts, and carefully select a large number of clause types worth human attention (Borchmann et al, 2020;Wang et al, 2021b;Hendrycks et al, 2021). Due to the repetition of contract language that new contracts usually follow the template of old contracts (Simonson et al, 2019), existing methods tend to incorporate structure information to tackle CCE. For example, Chalkidis et al (2017) assign a fixed extraction zone for each clause type and limit the clauses to be extracted from corresponding extraction zones.…”

Supervised Identification of Participant Slots in Contracts

“…Prior NLP work on contract language is extensive, including summarization (Manor and Li, 2019;Keymanesh et al, 2020) information extraction and understanding (Anish et al, 2019;Borchmann et al, 2020;Agarwal et al, 2021), as well as corpus studies looking at intrinsic properties of contracts (Curtotti and McCreath, 2011;Simonson et al, 2019) or providing new annotations over contract language (Funaki et al, 2020).…”

“…For coreference resolution, the system employs a heuristic technique. Since contract language tends to avoid 3rd person anaphora (Simonson et al, 2019), the system avoids some of the more challenging coreference problems and targets primarily noun phrases headed by common nouns and proper nouns. First, all 1st and 2nd person pronouns are joined into chains unique to each pronoun since their intrinsic deictic properties make them unambiguous.…”

“…Contracts contain language that is more repetitive than those typically studied in NLP research (Simonson et al, 2019), but this is not necessarily consistent language. A disclosing party may be referred to as a "Disclosing Party," "Discloser," "Providing Party," "Company," "Owner," "Client," or any of a limitless set of possible referring expressions, including the actual company name of the disclosing party.…”

Graph-based Keyword Planning for Legal Clause Generation from Topics

“…The fundamental units of discourse in contracts consist of "clauses" that are paragraphs of text that outline the terms and conditions of various types or topics (e.g., severability, benefits) (Table 1). Legal clauses can be characterized by their high inter-sentence similarity, and topic-specific content (Simonson et al, 2019). For example, Zhong et al (2020) showed that the sentences in legal corpora are almost 20% similar to each other.…”