--- title: "CCONJ: coordinating conjunction — definition and key papers" description: "CCONJ is the Universal Dependencies part-of-speech tag for coordinating conjunctions. Among them: Universal Dependencies v1: A…" canonical: "https://research.elamur.ai/free/en/concepts/upos-cconj/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/concepts/upos-cconj/) [home](https://md.research.elamur.ai/free/en/) / [concepts](https://md.research.elamur.ai/free/en/concepts/) / CCONJ: coordinating conjunction # 🏷️ CCONJ: coordinating conjunction Universal Dependencies · 7 key papers · [all concepts](https://md.research.elamur.ai/free/en/concepts/) Part of: [🗨️ Natural language processing](https://md.research.elamur.ai/free/en/concepts/natural-language-processing/) · [🌐 Universal Dependencies](https://md.research.elamur.ai/free/en/concepts/universal-dependencies/) CCONJ is the Universal Dependencies part-of-speech tag for coordinating conjunctions. These are words such as English "and", "or" and "but" that join two or more words, phrases or clauses of equal syntactic rank without making one depend on the other. Also: coordinating conjunction, coordinator, CONJ (older name), cc word ## What it is In [Universal Dependencies](https://md.research.elamur.ai/free/en/concepts/universal-dependencies/) (UD), every word of a sentence receives one of a closed list of seventeen universal part-of-speech tags ([Nivre et al. 2020](https://md.research.elamur.ai/free/en/papers/nivre-2020-universal-dependencies-v2-evergrowing-multilingual/)). CCONJ is the tag for coordinators. A coordinator links elements that stand side by side in the syntax, such as two nouns, two verbs or two clauses. It also signals how they relate: addition, alternative, contrast or exclusion. The defining test is the absence of subordination. A word that introduces a clause as a dependent of another clause is not a coordinator. "Because" in "she left because it rained" is one such word, and it belongs to SCONJ instead. The tag has a history. The universal tagset of Petrov, Das and McDonald proposed twelve coarse categories, and a single CONJ covered conjunctions there ([Petrov et al. 2012](https://md.research.elamur.ai/free/en/papers/petrov-2012-universal-part-speech-tagset/)). UD v1 extended that Google set to seventeen tags and gave subordinators their own category ([Nivre et al. 2016](https://md.research.elamur.ai/free/en/papers/nivre-2016-universal-dependencies-v1-multilingual-treebank/)). Version 2 of the guidelines then renamed the remaining CONJ to CCONJ, so the name now states its scope ([Nivre et al. 2020](https://md.research.elamur.ai/free/en/papers/nivre-2020-universal-dependencies-v2-evergrowing-multilingual/)). ## Examples Some invented sentences, with the coordinator marked by brackets: - English: "tea [and] coffee", "now [or] never", "small [but] loud" - French: "le pain [et] le vin", "[ni] lui [ni] moi", "il pleut, [mais] je sors" - German: "Hund [und] Katze", "[entweder] heute [oder] morgen" - Czech: "otec [a] syn", "chtěl, [ale] nemohl" In "either… or" and "ni… ni", both parts are usually tagged CCONJ. The first part may carry a subtype of the relation, such as English cc:preconj. ### Borderline cases | Neighbour | Typical words | Why not CCONJ | | --- | --- | --- | | SCONJ | because, if, that, French "que" | introduces a dependent clause | | ADV | however, therefore, French "pourtant" | moves freely in the clause; a discourse link, not a coordinator | | ADP | English "than", "with" in "bread with butter" | takes a nominal complement | | PART / DET | "both" in "both… and" | analysis varies by treebank; often still CCONJ in correlative use | Fixed expressions such as "as well as" or French "ainsi que" are built from words of other classes. They are joined by the fixed relation, and the individual tags depend on each treebank's conventions. ## Across languages Coordinators differ in form more than in function. Some are clitics or prefixes. Examples are Latin "-que", the Hebrew prefix ו "and", and Arabic "wa-". UD takes the syntactic word as its unit of analysis. Such elements are therefore split off as separate tokens, even where the spelling joins them to the next word ([Marneffe et al. 2021](https://md.research.elamur.ai/free/en/papers/marneffe-2021-universal-dependencies/); [Nivre et al. 2016](https://md.research.elamur.ai/free/en/papers/nivre-2016-universal-dependencies-v1-multilingual-treebank/)). This matters for [Latin](https://md.research.elamur.ai/free/en/topics/latin/) and [Hebrew & Aramaic](https://md.research.elamur.ai/free/en/topics/hebrew-aramaic/) corpora, where segmentation must come before tagging. Some languages lean on asyndetic coordination, with no coordinator at all. Others coordinate nouns with comitative or particle-like markers, which raises the choice between CCONJ and ADP or PART. Features are rare on this class. A coordinator usually has an empty FEATS column, since it does not inflect in most languages. Word-level features defined through Interset still apply where relevant, for instance foreign or abbreviated forms. Interset was developed as a shared feature inventory for converting between tagsets ([Zeman 2008](https://md.research.elamur.ai/free/en/papers/zeman-2008-reusable-tagset-conversion-tagset-drivers/)). ## In treebanks and research In a UD tree the conjuncts are linked to each other. The first conjunct is the head, and each later conjunct attaches to it by conj. The coordinator itself depends on the conjunct it introduces through the cc relation. UD v2 chose that conjunct, the one following the coordinator, over the first conjunct. This follows the general principle that function words hang from content words ([Marneffe et al. 2021](https://md.research.elamur.ai/free/en/papers/marneffe-2021-universal-dependencies/)). The enhanced graph can add further implied links, so that a shared subject reaches every conjunct ([Marneffe et al. 2021](https://md.research.elamur.ai/free/en/papers/marneffe-2021-universal-dependencies/)). The relations a CCONJ word takes in practice are dominated by cc in both English and French UD data. A thin tail of other relations remains: - conj, where the coordinator itself is coordinated, as in "and/or"; - reparandum, for speech disfluencies; - root, in sentence fragments; - advmod, for adverb-like uses; - fixed, in multiword coordinators, a usage that appears more in [French](https://md.research.elamur.ai/free/en/topics/french/). The tag also accounts for a few percent of all tokens in both languages. For [part-of-speech tagging](https://md.research.elamur.ai/free/en/concepts/part-of-speech-tagging/) and [dependency parsing](https://md.research.elamur.ai/free/en/concepts/dependency-parsing/), CCONJ is one of the easier categories, because the class is small and closed. Its edges still produce disagreements between treebanks of one language: - "so" and "yet" may be treated as coordinators or adverbs; - "for" may be treated as a coordinator or a subordinator; - correlative first parts are handled differently. Consistency checks bring such clashes to light. One approach compares UPOS trigram distributions across treebanks ([Aggarwal & Zeman 2020](https://md.research.elamur.ai/free/en/papers/aggarwal-2020-estimating-pos-annotation-consistency-different/)). Another flags annotations that differ for identical contexts ([Marneffe et al. 2017](https://md.research.elamur.ai/free/en/papers/marneffe-2017-assessing-annotation-consistency-universal-dependencies/)). Tutorials on UD principles treat coordination as a central case of how the scheme separates content words from function words (Zeman 2021). Coordination is also hard for parsers beyond tagging. The coordinator is easy to identify, but deciding which spans it joins is not. See [Treebanks](https://md.research.elamur.ai/free/en/topics/treebanks/) and [Morphology](https://md.research.elamur.ai/free/en/topics/morphology/) for related material. How often the tag CCONJ occurs: its share of all tokens Counted by code over the Universal Dependencies 2.18 treebanks of English and French (every token once, subtypes with their base relation). **The numbers** | Treebanks | Share | | --- | --- | | English treebanks | 2.71% of tokens (30,532 tokens) | | French treebanks | 3.78% of tokens (41,249 tokens) | The relations CCONJ words take most often The five most frequent in each language's treebanks, by tokens. **The numbers** | Pair | Tokens | | --- | --- | | English: cc | 30,224 | | English: conj | 119 | | English: reparandum | 90 | | English: root | 33 | | English: advmod | 26 | | French: cc | 40,744 | | French: fixed | 129 | | French: reparandum | 98 | | French: advmod | 85 | | French: conj | 66 | ## Key papers The works this page cites, the most cited first; citations as counted by OpenAlex plus the papers of this library that cite the work, read 2026-10-24. - [Universal Dependencies v1: A Multilingual Treebank Collection](https://md.research.elamur.ai/free/en/papers/nivre-2016-universal-dependencies-v1-multilingual-treebank/) — Joakim Nivre, Marie-Catherine de Marneffe, Filip Ginter et al. · 2016 · 1,741 citations - [A Universal Part-of-Speech Tagset](https://md.research.elamur.ai/free/en/papers/petrov-2012-universal-part-speech-tagset/) — Slav Petrov, Dipanjan Das, Ryan McDonald · 2012 · 1,180 citations - [Universal Dependencies](https://md.research.elamur.ai/free/en/papers/marneffe-2021-universal-dependencies/) — Marie-Catherine de Marneffe, Christopher D. Manning, Joakim Nivre et al. · 2021 · 480 citations - [Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection](https://md.research.elamur.ai/free/en/papers/nivre-2020-universal-dependencies-v2-evergrowing-multilingual/) — Joakim Nivre, Marie-Catherine de Marneffe, Filip Ginter et al. · 2020 · 325 citations - [Reusable Tagset Conversion Using Tagset Drivers](https://md.research.elamur.ai/free/en/papers/zeman-2008-reusable-tagset-conversion-tagset-drivers/) — Daniel Zeman · 2008 · 181 citations - [Assessing the Annotation Consistency of the Universal Dependencies Corpora](https://md.research.elamur.ai/free/en/papers/marneffe-2017-assessing-annotation-consistency-universal-dependencies/) — Marie-Catherine de Marneffe, Matias Grioni, Jenna Kanerva et al. · 2017 · 30 citations - [Estimating POS Annotation Consistency of Different Treebanks in a Language](https://md.research.elamur.ai/free/en/papers/aggarwal-2020-estimating-pos-annotation-consistency-different/) — Akshay Aggarwal, Daniel Zeman · 2020 · 7 citations ## Related concepts - [Universal Dependencies](https://md.research.elamur.ai/free/en/concepts/universal-dependencies/) — Universal Dependencies is a framework and an open collection of treebanks that annotates the grammar of many languages with one shared… - [Part-of-speech tagging](https://md.research.elamur.ai/free/en/concepts/part-of-speech-tagging/) — Part-of-speech tagging is the task of assigning each token in a text a grammatical category, such as noun, verb, adjective or determiner… - [Morphology](https://md.research.elamur.ai/free/en/concepts/morphology/) — Morphology is the branch of linguistics that studies the internal structure of words. It covers how words are built from meaningful units… ## More papers Topics: [🔤 Part-of-speech tagging](https://md.research.elamur.ai/free/en/topics/part-of-speech-tagging/) · [🌐 Universal Dependencies](https://md.research.elamur.ai/free/en/topics/universal-dependencies/) Shelves: [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) - [← previous concept ccomp: clausal complement](https://md.research.elamur.ai/free/en/concepts/ud-ccomp/) - [next concept → Chatbots](https://md.research.elamur.ai/free/en/concepts/chatbot/) - [↑ back to concepts](https://md.research.elamur.ai/free/en/concepts/)