Skip to navigation – Site map

HomeCall for papersCall for papers – openN°26 – Re-thinking frozenness and...

N°26 – Re-thinking frozenness and variability

Deadline for sending in final versions of papers: 30 September 2025

Lexis Journal in English Lexicology – will publish its 26th issue in 2025. It will be edited by Vincent Hugou (Sorbonne Université, France) and Mathilde Pinson (Université Sorbonne Nouvelle, France) and will deal with the topic “Re-thinking frozenness and variability”.

Re-thinking frozenness and variability

The definition of a frozen sequence can be limited to a phrasal polylexical expression (e.g. a proverb), or an infra-phrasal expression (e.g. a nominal compound, a phrase), which is identifiable by its morpho-syntactic fixedness, its non-compositionality and because it is often also the place where cultural representations can be found. The importance of frozen sequences in language, compared with free sequences (according to Erman & Warren [2000], there are more than 50% of prefabricated expressions in oral and written discourse), underlines the fact that this is by no means a marginal phenomenon, as is abundantly clear from the many studies on the subject (e.g. Gross [1988], Mejri [2013], for French; Brinton & Traugott [2005], for English, etc.). The ‘deformation’ of an expression is traditionally seen as the normal counterpart of ‘frozenness’. In fact, frozenness is obviously a necessary pre-requisite for deformation, i.e. the introduction of variability within a frozen expression, thereby entailing the transgression of certain syntactic restrictions and, prototypically, a return to a literal meaning. A frozen sequence would thus pertain to langue (language), and its manipulated counterpart to parole (speech/discourse).

However, adopting a traditional definition of frozen sequences and their deformation may give an overly simplistic view of the phenomenon under study. The 26th issue of Lexis offers to take a fresh look at these two concepts.

Frozen sequences and their transformed versions constantly interact with each other. Most studies on frozenness agree that there is a continuum between free sequences and frozen sequences, and that the criteria for recognising a frozen expression often have to be determined on a case-by-case basis. For example, for French, Guimier & Oueslati [2006] set up a battery of ad hoc tests to assess the degree of frozenness of a family of ‘verb+invariant adjective’ constructions (for example, boire sec, tourner court). They also show that certain occurrences exist in both fixed and non-fixed forms, depending on the context (jouer subtil = to play subtly or (ne pas) jouer subtil = (not to) beat around the bush).

In addition, a frozen expression that has been subject to variability in the course of discourse production may eventually become a new frozen expression, which will enter the linguistic system at some point; it would then be possible to speak of a ‘re-formed’ sequence, which may be deformed again, and so on. Therefore, a diachronic study makes it possible to understand how an expression has become frozen as a chunk over the course of time (see, for example, Traugott [2020]). However, a frozen expression may also be creatively deformed from time to time, only for playful purposes (see Moon [1988]). Nevertheless, it remains impossible to state definitively that what has been transformed in discourse will not later be exploited as the basis for a new re-formation. For example, the frozen expression worn out was deformed when nouns were inserted in place of worn - at first mainly nouns linked to an activity (shopped out), then nouns denoting humans (peopled out) and substances (sugared out), etc. It then gradually clustered around exemplars, i.e. productive and interpretative models. For example, coffeed out is a conventionalised exemplar, which now has its own entry in dictionaries. This form undoubtedly serves as a basis for other semantically related forms: ice-teaed out, energy drinked out, etc. (Hugou [2013]).

Although the two notions of frozenness and deformation/variability are often thought of as a living and constantly evolving whole, some expressions are less prone to deformation than others. For example, the phrase all of a sudden is subject to blocking on both the syntagmatic and paradigmatic axes; this phenomenon could be described as ‘hyper-frozenness’ or ‘absolute frozenness’.

The delimitation of the notions of frozenness and deformation also raises some questions. Frozenness is traditionally determined according to a number of applicable criteria. For example, Benson [1985] places collocations halfway between frozen expressions and free combinations, on the grounds that they constitute transparent semantic groupings (for example, high altitude, high wall). However, in high wind or high hopes the collocative high is used metaphorically in relation to the other element, and the combinations are partially compositional. In this sense, they are frozen (or at least semi-frozen) sequences. Villers [2010] shares the same view regarding proverbs, which have often been excluded from frozen sequences, since many of them have a literal interpretation (for example, first come, first served). It is therefore important to bear in mind the scalar nature of frozenness, both formally and semantically. Note that the prosodic criterion can also be added to the clues for recognising a fixed expression, as some expressions have a specific intonational contour (see Lacheret & Legallois [2013], Crystal [2019]).

As a result, it should be possible to posit degrees of deformation, just as there are degrees of frozenness. When does deformation start and where does it end? According to the traditional definition provided in the introduction, deformation can be assimilated to a transformation, or even a manipulation of a frozen sequence, which makes it possible to return to a free combination of words and a literal meaning. But the question is not so simple. In terms of form, it is generally accepted that a frozen sequence cannot undergo lexical insertions. There are exceptions to this principle: for example, the insertion of an adjective (he made some serious coin vs. the original he made coin), or an adverb with a modal value (he made some really serious coin), does not affect the internal structure of the frozen sequence, and therefore does not constitute evidence of any deformation. Having said that, on closer inspection, the addition of an adjective like quick in I’m going to go grab a quick forty winks also triggers, at the same time, the use of the indefinite article a, which was not present in the expression used as a basis, to grab forty winks. Could we still talk about deformation, since the stability of the initial form has been disrupted? Besides, can deformation take place without the slightest alteration to the surface signifier, simply by creatively extending it? This applies to the established expression over the moon in I guess now you are over the moon, Mars, Jupiter and the whole galaxy (example cited in Carter [2015]). However, there may be invariant elements that cannot be changed. Just think of the binary rhythm of many proverbs, which is a strong clue as to the recognition of paremiological expressions.

As far as meaning is concerned, the question of the limits of deformation also arises in the case of simple synonym substitution (paradigmatic blocking being a criterion of frozenness), but without there being a return to the literal meaning. For example, in happy as a pig in mud/slop/shit/poo(p), the paradigm is open and the speaker ‘plays’ with near-synonyms. Is this still a case of deformation if the criterion of a return to compositional meaning is not met? The situation is somewhat different in force/elbow/shoulder/fight/cheat/cry your way through something. There is no synonym substitution as in the previous vivid simile, where the semantic relationship is stronger. Israel [1996] has shown that the first instantiations of the way-construction emerged with verbs of movement on the model of wend your way, and they then served as exemplars. However, to be perfectly realistic, it is difficult to believe that today’s speakers keep the occurrence wend your way in mind when they produce Rasselas dug his way out of the Happy Valley (example from Israel, ibid.). Or could they be working on a partly schematic pattern, for example ‘X one’s way through Y’? And in this case, if there is no specific underlying form, is the notion of ‘deformation’ still relevant?

In other words, is the act of ‘deformation’ solely dedicated to playful transformations (see Milner’s ‘syntactic puns’ [1978], or Grunig’s exploitation of deformation in media productions, such as advertising discourse [1998]), or should a broader definition be adopted?

Freezing and deformation can also be examined from the point of view of the construction of meaning. It is generally accepted that the meaning of a frozen sequence is grasped as a whole, but it seems possible (and not contradictory) to argue that each of the constituent parts, whether free or bound morphemes, still has a role to play in the meaning of the expression, and even in its pragmatic value. What’s more, some frozen sequences contain ‘fossilised’ elements, which are often evidence of an older state of the language. The proverb there’s many a slip twixt cup and lip often appears, especially in American English, in the form there’s many a slip between the cup and the lip, as does the established formula till death us do part, which is often transformed into a slightly more canonical form (SVO order), till death do us part. If one assumes that these are deformations in a broad sense of the notion of deformation (see the limits of deformation above), and although there is no return to a literal meaning, it has to be said that the speaker can sometimes choose to transmute a form for reasons other than playfulness. In the case of this proverb and this religious formula, deformation and remotivation go hand in hand.

More generally, the question of how a frozen expression is reactivated or brought back to the consciousness of the person who has to decode it is worth asking. How many ‘clues’ are needed to recognise the underlying message/the sub-text? What role should the cotext and/or context play? Psycholinguistic experiments have shown that these reactivations are all the easier when they occur at the end of frozen expressions (Denhière & Verstiggel [1997: 139]).

Finally, is the notion of deformation still relevant if the communicative intention of the speaker is not perceived? The double reading that should be triggered by the name of an Asian food restaurant, Absolutely Phobulous, is lost without any knowledge of the sub-text, Absolutely Fabulous, a British TV series that was in vogue about 25 years ago.

Finally, the study of frozenness and deformation raises questions of a terminological and, consequently, conceptual nature. Tournier & Tournier [2009: 151] think that these concepts are non-operational, if not unsatisfactory, and would best be replaced by ‘lexicalisation’. Frozenness and deformation also conjure up a large number of other related notions.

  • In the case of frozen sequences, these include chunks, constructions (in the sense of Construction Grammars, see Goldberg [1995], [2006]), formulas, idiomaticity, lexicalisation, locutionality, phraseological units, prefabricated units, etc.

  • Regarding deformation, notions such as anti-lexicalisation (Brinton & Traugott [2005]), creativity (as defined by Bauer [1983]), delexicalisation, deautomatisation (Gonzáles-Rey [2015]), transgression, recompositionalisation (Pinson [2023]) can also be mentioned.

It may then be worth determining, as Mejri [2006] has already done, for example, whether all these terms can be superimposed on frozenness and deformation, or whether each of them actually refers to finer distinctions. To take just two examples, if the concept of ‘lexicalisation’ is taken to mean the establishment of an expression in the repertoire of a community, then frozenness can be equated with lexicalisation (a position supported, for example, by Lehmann & Martin-Berthet [2008: 222]). By contrast, if the term is taken in the sense of the transformation of a grammatical expression into a lexical expression (as in Brinton & Traugott [2005]), then the two concepts are not synonymous. A nursery rhyme (eenie meenie miney mo, catch a tiger by the toe, etc.) is formally, semantically and culturally frozen, but is not lexical in the narrow sense of the term. There is also reason to wonder whether a construction, in the sense of Construction Grammars, refers to the same linguistic object as a frozen expression.

These paragraphs reveal that the field of frozenness and deformation is alive and well, replete with issues (in particular because of other neighbouring notions), and principles to be further explored (such as a better understanding of the mechanisms of deformation/variability). The following are non-exclusive and non-exhaustive suggestions:

  • Theoretical framework and frozenness/deformation: what theories are best able to capture these phenomena?

  • A comparison of related concepts (lexicalisation, etc.) with frozenness and deformation: similarities, differences.

  • The limits of the field under study: where do frozenness and deformation begin and end?

  • Recognition of a frozen or deformed expression: is there a ‘grammar’ of deformation (i.e. a phenomenon that is not devoid of systematicity), or does deformation occur without any definite plan on the part of the speaker? Are certain expressions more apt to be deformed? The question of cotextual elements favouring deformation could also be raised.

  • Frozenness/deformation and meaning calculation: clues for recognising the sub-text, role of the cotext and/or context.

  • Pragmatic functions of deformation. Deformation and discourse genres: advertising and political discourse (slogans, names of political parties); humour and frozenness.

Systematic analyses, based on concrete examples (a family of idioms for example) and/or which attempt to clarify the notions of frozenness and deformation, or which discuss the notions under study together with related notions (lexicalisation, etc.) will be welcome. A contrastive approach between English and other languages can also help determine whether all languages are governed by the same mechanisms.


Bauer, Laurie. 1983. English Word-Formation. Cambridge: Cambridge University Press.

Benson, Morton. 1985. Lexical combinability. Papers in Linguistics 18. 3-15.

Brinton, Laurel & Traugott, Elizabeth Closs. 2005. Lexicalization and Language Change. Cambridge: Cambridge University Press.

Carter, Ronald. 2015. Language and Creativity: The Art of Common Talk. Oxon: Routledge.

Crystal, David. 2019. Sounds Appealing: The Passionate Story of English Pronunciation (chapter 6). London: Profile Books.

Denhière, Guy & Verstiggel, Jean-Claude. 1997. Le traitement cognitif des expressions idiomatiques. Activités automatiques et délibérées. In Pierre Fiala, Pierre Lafon & Marie-France Poguet (eds.), La locution : entre lexique, syntaxe et pragmatique. Identification en corpus, traitement, apprentissage, 121-138. Paris: Klincksieck, Collection Saint-Cloud.

Erman, Britt & Warren, Beatrice. 2000. The idiom principle and the open choice principle. Text 20. 29-62.

Goldberg, Adele. 1995. Constructions: A Construction Grammar Approach to Argument Structure. Chicago/London: University of Chicago Press.

Goldberg, Adele. 2006. Constructions at Work: The Nature of Generalization in Language. Oxford: Oxford University Press.

Gonzales-Rey, Isabel. 2015. La Phraséologie du français. Toulouse: Presses Universitaires du Midi.

Gross, Gaston. 1996. Les Expressions figées en français. Paris: Ophrys.

Grunig, Blanche-Noëlle. 1998. Les Mots de la publicité. Paris: CNRS-Éditions.

Guimier, Claude & Oueslati, Lassaâd. 2006. Le degré de figement des constructions ‘verbe+adjectif invarié’. In Jacques Francois & Salah Mejri (eds.), Composition syntaxique et figement lexical, 17-38. Caen: Presses Universitaires de Caen.

Hugou, Vincent. 2013. Productivité et émergence du sens. L’exemple de la construction (all) X-ed out dans un corpus de blogs et de forums de discussion. PhD diss., Paris 3, Sorbonne Nouvelle.

Israel, Michael. 1996. The Way Constructions Grow. In Adele Goldberg (ed.), Conceptual Structure, Discourse and Language, 217-230. Stanford: CSLI.

Lacheret, Anne & Legallois, Dominique. 2013. Expressivité vocale et grammaire : comment le symbolique construit le prosodique. In Martine de Gaudemar (ed.), Les plis de la voix, 45-56. Limoges: Lambert-Lucas.

Lehmann, Alise & Martin-Berthet, Françoise. 2008. Introduction à la Lexicologie: Sémantique et Morphologie. Paris: Armand Colin.

Mejri, Salah. 2006. La terminologie du figement : approche contrastive (français-arabe). Syntaxe & Sémantique 7. 139-152.

Mejri, Salah. 2013. Figement et défigement : problématique théorique. Pratiques. 79-97.

Milner, Jean-Claude. 1978. De la syntaxe à l’interprétation : Quantités, insultes, exclamations. Paris: Seuil.

Moon, Rosamund. 1998. Fixed Expressions and Idioms in English: A Corpus-based Approach. Oxford: Oxford University Press.

Pinson, Mathilde. 2023. Decompositionalization and partial recompositionalization: The emergence of by the same token as a polyfunctional discourse marker. Journal of English Linguistics 51(3). 236-264.

Tournier, Nicole & Tournier, Jean. 2009. Dictionnaire de Lexicologie Française. Paris: Ellipses.

Traugott, Elizabeth Closs. 2020. The development of ‘digressive’ discourse-topic shift markers in English. Journal of Pragmatics 156. 121-135.

Villers, Damien. 2010. Les modalités du détournement proverbial : entre contraintes et libertés. Modèles linguistiques 62.

How to submit

Please clearly indicate the title of the paper and include an abstract between 3,000 and 6,000 characters (including spaces) as well as a list of relevant key-words and references. All abstract and paper submissions will be anonymously peer-reviewed (double-blind peer reviewing) by an international scientific committee composed of specialists in their fields. Papers will be written preferably in English or occasionally in French.

Manuscripts may be rejected, accepted subject to revision, or accepted as such. There is no limit to the number of pages.

Submissions (abstracts and articles, in Word document) will be submitted via the journal’s submission platform. If you encounter any problem, please send a message to Lexis.


June 2024: Call for papers

October 31 2024: Deadline for submitting abstracts to Lexis via the journal’s submission platform

January 2025: Evaluation Committee’s decisions notified to authors

May 30 2025: Deadline for submitting papers via the journal’s submission platform (Guidelines for submitting articles:

June and July 2025: Proofreading of papers by the Evaluation committee

August 1 2025 to September 2025: Authors’ corrections

September 30 2025: Deadline for sending in final versions of papers

Search OpenEdition Search

You will be redirected to OpenEdition Search