Wissenschaftlicher Sammelband, herausgegeben von Thomas Tinnefeld unter Mitarbeit von Ines-A. Busch-Lauer, Hans Giessen, Michael Langner, Adelheid Schumann. Saarbrücken: htw saar 2012. ISBN 978-3-942949-00-2.

Formulaic Expressions for Foreign Language Learning

Günter Schmale (Metz, France)

Abstract (English)
In consideration of long-term linguistic insight into the existence of language formulaicity, of late supported by quantitative studies on the high degree of formulaic expressions in different text types, foreign language methodology cannot but reflect on the role of prefabs for foreign language teaching. Following the reflexion on learner adequate formulaic expressions, one specific type, viz formal idioms, will be discussed. Formal idioms are particularly useful for foreign language learning because of their creative potential. In consideration of methodological principles for their acquisition, specific types of “construction” will be developed for the use of different functions of the German verbs müssen and sollen, considering that morphosyntactic “atomistic” rules for their use have not proved to be sufficient and efficient.
Key words: formulaic expressions, foreign language methodology, the German verbs müssen and sollen

Abstract (Deutsch)
Angesichts teilweise sehr alter Erkenntnisse zur Präformierung von Sprache, in neuerer Zeit durch quantitative Ergebnisse zur häufigen Verwendung vorgeformter Ausdrücke untermauert, muss die FS-Didaktik über die Rolle vorgeformter Ausdrücke beim FS-Erwerb reflektieren. Im Anschluss an eine Diskussion für das FS-Lernen relevanter präformierter Konstruktionseinheiten wird ein Typ analysiert - formale Idiome oder Konstruktionen - der aufgrund ihrer Ausdrucksflexibilität für den FS-Lerner hochgradig relevant ist. Derartige “Konstruktionen“ werden – unter Berücksichtigung didaktischer Prämissen – für unterschiedliche Funktionen der deutschen Modalverben bzw. modalisierenden Verben müssen und sollen entwickelt, für deren adäquaten Gebrauch „atomistische“ Grammatikregeln nicht ausreichend sind.
Stichwörter:    präformierte Konstruktionseinheiten, Fremdsprachendidaktik, Modalverben müssen und sollen

1   Introduction – The importance of Formulaic Expressions in Language Production

Our language does not expect us to build everything starting with lumber, nails, and blueprint, but provides us with an incredibly large number of prefabs, which have the magical property of persisting even when we knock some of them apart and put them together in unpredictable ways. (Bolinger 1976: 1)

Dwight Bolinger, in the aforementioned passage from his wonderful treaty of Meaning and Memory (1976), was not the first linguist to point out the existence of “prefabs” or pre-fabricated expressions. In fact, a great number of linguists[1] did so long before him, e.g. Bréal (1897) mentioning “groupes articulés”[2], Bally (1909) “groupements usuels”, Saussure (1916) “locutions toutes faites”, Sapir (1921) “compounded elements”, Bloomfield (1933) qualifying the “lexicon as a list of basic irregularities”, Jespersen (1934) referring to “formulae”, Porzig (1934) to “wesenhafte Bedeutungsbeziehungen”, not to forget Firth’s (1937) seminal work on collocations[3]. More recently, Coseriu (1967) comments on “lexikalische Solidaritäten”, while Mel’čuk 1998 states that “People don’t speak in words, they speak in phrasemes.” (Mel’čuk 1998: 169) and Mejri observes that “Frozenness is a process that is inherent in all modern languages.” (Mejri 2007: 685) – and last but not least the “idiomatic footprint”[4] left in any language production suggested by Feilke (1998). Gasparov even goes as far as to affirm:

Whatever we say or perceive in speech is made from other facts of speech, which we recognize more or less as being present in our previous experience and being set in our memory. (Gasparov (2004): 46)[5]

Gasparov’s assumption is supported by a number of recent studies on the quantitative presence of formulaic expressions in “adult native language”, for instance the one by Wray & Perkins (2000):

If we take formulaicity to encompass, as some do, also the enormous set of ‘simple’ lexical collocations, […], then possibly as much as 70% of our adult native language may be formu­laic […]. A range of corpus studies […] have shown that the patterning of words and phrases in ordinary language manifests far less variability than could be predicted on the basis of grammar and lexicon alone, and in fact most natural language, written or spoken, appears to consist largely of collocational ‘sets’ or ‘frameworks’ […].” (Wray & Perkins 2000: 1-2)

Referring to Altenberg (1998), Wray even assumes that “formulaic material in normal languagemight stretch “as high as 80 per cent in normal language” (Wray 2000: 466), whereas Erman & Warren (2000) find 52% of formulaic expres­sions in written and 58% in oral language production; in a study of the scenario of “Some like it hot”, van Lancker-Sidtis & Rallon discover that 25% of the “sentences” contain speech formulas, idioms, proverbs and other formulaic expressions (Lancker-Sidtis & Rallon 2004: 207). Obviously, further analyses of authentic corpora will be necessary, considering Donalies’ observation that the quoted quantitative studies are based on confusing criteria of what is formulaic (Donalies’ 2009: 29)[6]; there is nevertheless no doubt whatsoever that formulaic language plays an important role in native speaker language production and has, thus, to be taken into consideration by foreign language methodology in order to optimize foreign language learning (FLL).

Surprisingly, in spite of an otherwise stringent communicative orien­tation, the Common European Framework of Reference for Language (CEFR) (2001) does not consider knowledge of formulaic language as part of “grammatical competence”:

Grammatical competence may be defined as knowledge of, and ability to use, the grammatical resources of a language. Formally, the grammar of a language may be seen as the set of principles governing the assembly of elements into meaningful labelled and bracketed strings (sentences). Grammatical competence is the ability to understand and express meaning by producing and recognizing well-formed phrases and sentences in accordance with these principles (as opposed to memorizing and reproducing them as fixed formulae). (CEFR 2001: 112-113)

In fact, according to the CEFR, “grammatical competence” does not consist in the production of communicatively successful utterances, but in the generation and understanding of “well-formed phrases and sentences” in which “the grammatical resources of a language” are respected. It is this analytic construction of sentences “bottom up”, assembling morphemes, lexemes, phrases to create sentences, which, for the authors of the CEFR, seems to constitute the genuinely creative production of meaning, whereas the use of “fixed formulae” seems to be considered as a non-creative act of reproducing memorized prefabricated elements. Within the domain of “lexical competence”, however, the CEFR mentions different types of “fixed expressions”, “phrasal idioms” and “fixed collocations” etc. (CEFR: pp. 110), but without tempting to differentiate different types considering their FLL-relevance, and appreciating their communicative value in language production.

What is more, the CEFR perspective towards creative – analytic – language production is contrary to Henry Widdowson’s credo who, as early as 1989, emphasized the fact that “communicative competence” is by no means solely a question of knowing rules and producing utterances “from scratch”:

Communicative competence is not a matter of knowing rules for the composition of sentences and being able to employ such rules to assemble expressions from scratch as and when occasion requires. It is much more a matter of knowing a stock of partially pre-assembled patterns, formulaic frameworks, and a kit of rules, so to speak, and being able to apply the rules to make whatever adjustments are necessary according to contextual demands. (Widdowson’ 1989: 135)[7]

Around one decade later, Wray & Perkins (2000) or Aguado (2002) mentioned the impor­tance of formulaic sequences for foreign language acquisition. And, of course, “phraseodidactics”[8], a discipline developed mainly in German-speaking countries[9], researchers have been reflecting on different aspects of teaching phraseological expressions since the 1970s (Hallsteinsdottir 2011: 1). 

The objective of the present paper is to delimit and describe formulaic expressions (FE) from a methodological perspective, concentrating on one particular type of FE which will be illustrated with an example from German grammar.

The following section of this paper will deal with the question of what types of FE are relevant for FLL. Section 3 is dedicated to one type of formulaic expression widely discussed of late, i.e. “constructions”. Section 4 is dedicated to the presentation of difficulties learners encounter with the use of the German verbs müssen and sollen which will then lead to the description of basic constructions and principles for the learning and teaching of these verbs (chap. 5).  

2    Formulaic Sequences for Foreign Language Learning 

Taking Burger, Buhofer & Sialm’s (1982) and Burger’s (20104) classification of basic phraseological categories into sentential (proverbs, common places) and syntagmatic referential (idiomatic, partially idiomatic, collocations)[10] and communicative phrasemes as a starting point, the present section will delimit the formulaic or phraseological expressions or sequences which should be acquired by the foreign language learner in order to become part of his active communicative competence, but also which FE should be excluded from the learner’s active competence. Obviously the FEs used in texts or situations treated in the language classroom will have to be explained in order to become part of the learner’s passive or receptive competence. However, in no case should context-free lists of phraseological expressions constituting a phraseological minimum or optimum be distributed to students (Schmale 2009).

Three main types of FEs are an indispensable part of native speaker but also non-native speaker competence:
  • Communicative phrasemes (Burger 20104) - also called pragmatic idioms (Burger 1973) or routine formulae (Coulmas 1981) - which serve the accomplishment of stereotyped speech acts closely linked to specific communicative situations, e.g. good morning, excuse me, congratulations in English and their French and German equivents bonjour / Guten Morgen, excusez-moi / Entschuldigung, félicitations / Herzlichen Glückwunsch.
  • Collocations or “lexical solidarities”: more or less stable or fixed polylexical combinations of lexemes whose meaning is semantically compositional[11], e.g. to brush one’s teeth[12], to give / offer someone a lift in English; zu Ende gehen, etwas zur Folge haben in German; attendre avec impatience or changer d’avis in French.
  • Some types of partially idiomatic expressions which are fully lexicalized and do not have a completely non idiomatic equivalent, for instance the German blinder Passagier (a stowaway), who is not blind at all[13], but rather clandestine as French has it in its passager clandestin; a blind date or a nervous cough in English, where the respective adjectives do not directly qualify the noun, but, often in a metaphorical way, the agent involved in the action designated by the substantive. Compared to the two – very large – preceding classes of routine formulae and collocations, this is however a minor category, comprising to a large extent what Burger (20104: pp. 49) refers to as phraseological terms[14] most often belonging to a specialized language from the domain of economy, law, medicine or philosophy.

On the other hand, there are strong arguments for the exclusion of other types of formulaicity from active non-native speaker competence:
  • proverbs such as All that glitters is not gold[15] in English and common places like Aus den Augen, aus dem Sinn in German[16];
  • (partially) syntagmatic idiomatic expressions containing a metaphor and / or an image such as to drop a brick or as drunk as a lord / skunk[17] in English; combat / échange à fleuret moucheté or fier comme Artaban in French; jemandem einen Korb geben or dumm wie Bohnenstroh in German;
  • finally, certain rare types of routine formulae whose use by non-native speakers might imply the taking of a position haute, a conversationally superior position and, thus, be considered as a potentially face-threatening act, for instance in the domain of “defending the turn” via a formula such as the French je termine, the German lassen Sie mich bitte ausreden or the English hang on a second, which could not be tolerated when uttered by a non-native speaker, unless he or she disposes of a highly developed near-native competence.

The following arguments substantiate the exclusion of the three aforementioned types of phrasemes from active communicative non-native speaker competence:

(1) First and foremost, unlike most routine formulae and collocations (and some partially idiomatic phraseological expressions), which do not have non-phraseological equivalents, proverbs, commonplaces and idiomatic expressions do have non-idiomatic counterparts expressing the same semantic contents[18].

(2) Competent and adequate use of idiomatic sentential and syntagmatic presupposes knowledge and control of a large number of various conditions concerning style, social status and relation, situation, or register, which are extremely difficult to acquire in an institutional context, i.e. in non real-life situations. Their use thus implies a strong danger of inadequate use.

(3) According to Dobrovol’skij & Lûbimova (1993) the – even perfectly correct – use of highly idiomatic metaphorical and / or imagery expressions by non-native speakers is often negatively sanctioned by native speakers.[19]

Als Nichtmuttersprachler muss man sozusagen immer ein doppeltes Spiel spielen nach dem Prinzip: Ich fühle mich zwar in dieser Kultur wie zu Hause, bin mir aber ständig darüber im Klaren, dass es sich für mich dabei um eine fremde Kultur handelt. (Dobrovol’skij & Lûbimova 1993: 156)

In fact, in spite of a possibly developed proficiency in the foreign language, non-native speakers do not seem to be expected to use strongly idiomatic expressions as they are considered as an exaggerated assimilation of a culture which is not theirs[20]. This is especially the case when non-native competence is otherwise marked by a foreign accents or lexical, syntactic, pragmatics flaws which make the non-native speaker easily recognizable as a foreigner.

(4) Wray (2000) excludes idiomatic expressions based on non-compositional semantics, from foreign language learning because the learner cannot derive productive rules from them.

Word-strings presented in a syllabus must, if they are to enable the learner to infer lexical patterns or grammatical rules, be semantically and grammatically regular. It follows that some formulaic sequences, namely those that are non-canonical, metaphorical or archaic, must be excluded.” (Wray 2000: 482)

Some reservations have to be made, however, about this hypothesis:
  • Absolutely indispensable FEs such as routine formulae are, from a systematic point of view, far from being “grammatically regular”, e.g. the noun phrase good morning might be an elliptical form of I wish you a good morning, but this is by no means a productive model allowing, for instance, the creation of new car instead of I bought a new car;
  • Wray & Perkins’ (2000) distinction between “rule based” and thus creative production on the one hand and “holistic” and thus non creative reproduction on the other hand is not viable: firstly, creativity is not a valid criterion for expressions which do not possess a non-phraseological semantic equivalent; secondly, Gülich (2008) points out that the recourse to prefabricated language material is also to be considered as a productive or creative act as it has to be adequately deployed in communicative contexts; thirdly, it seems that the common cliché of idiomatic expressions as used, stereotyped, hackneyed language might have motivated Wray & Perkins view.


3    Constructions as Formulaic Expressions

3.1   Fillmore, Kay & O‘Connor’s (1988) Approach

Fillmore, Kay & O‘Connor’s (1988) widely discussed treatise of Regularity and Idiomaticity in Grammatical Constructions is based on the observation[21] that there are phenomena in language which cannot be explained via well-known grammatical theories.

As useful and powerful as the atomistic schema is for the description of linguistic competence, it doesn’t allow the grammarian to account for absolutely everything in its terms. […], the descriptive linguist needs to append to this maximally general machinery certain kinds of special knowledge – knowledge that will account for speakers’ ability to construct and understand phrases and expressions in their language which are not covered by the grammar, the lexicon, and the principles of compositional semantics, as these are familiarly conceived. Such a list of exceptional phenomena contains things which are larger than words, which are like words in that they have to be learned separately as individual whole facts about pieces of the language, but which also have grammatical structure, structure of the kind that we ordinarily interpret by appealing to the operation of the general grammatical rules. This list is not merely a supplement to the lexicon: it contains information about fully productive grammatical patterns, including what have been variously referred to as ‘minor sentence types’, ‘special constructions’, and the like. This ‘Appendix to the Grammar’ can be thought of as the repository of what is IDIOMATIC in the language.” (Fillmore, Kay & O’Connor 1988: 504)

More specifically, the authors define idiomaticity by three main criteria:

We think of a locution or manner of speaking as idiomatic if it is assigned an interpretation by the speech community but if somebody who merely knew the grammar and the vocabulary of the language could not, by virtue of that knowledge alone, know (i) how to say it, or (ii) what it means, or (iii) whether it is a conventional thing to say. Put differently, an idiomatic expression or construction is something a language user could fail to know while knowing everything else in the language.” (Fillmore, Kay & O’Connor 1988: 504)

Idiomatic “exceptional phenomena, minor sentence types, special constructions, locutions, manners of speaking (ibid.) as the quoted authors choose to call them cannot be explained by rules of generative grammar, but need to be specified by “not only syntactic, but also lexical, semantic, and pragmatic information; […].” (Fillmore, Kay & O‘Connor 1988: 501)[22] These “idiomatic constructions” attract Fillmore, Kay & O’Connor’s attention because they are “productive” and “highly structured” – two types are distinguished:
  • “Substantive idioms” whose “lexical make-up is (more or less) fully specified” (Fillmore, Kay & O‘Connor 1988: 505) and which correspond by and large to what classical phraseology has described so far (e.g. Burger 20104) as referential, communicative or structural phraseological expressions.
  • “Formal idioms” which “are syntactic patterns dedicated to semantic and pragmatic purposes not knowable from their form alone” (Fillmore, Kay & O‘Connor 1988: 505)[23]; these patterns are “lexically open” (ibid.).[24] Fillmore, Kay & O’Connor (discuss the famous “let alone” construction, e.g. “He doesn’t get up for lunch, let alone breakfast.” (id.: 517) or “Him be X”, eg. “Him be a doctor?”; (Kay & Fillmore 1997: 9) analyze the construction “What’s X doing Y?”. It is true that both Burger, Buhofer & Sialm (1982) and Fleischer (1982)[25] have mentioned this type of syntactic “frame” as “Modellbildungen” (models) or “Phraseoschablonen” (phraseological moulds) within the domain of phraseological research[26], but, unlike Fillmore, Kay & O’Connor (1988), not systematically taking into account both syntactic-semantic and pragmatic components of the respective construction. It is to be noted that the latter’s “pragmatic components” do refer neither to prosodic nor to non-verbal features of communication which definitely play a non-negligible role in multimodal communication in general, and in the production of formulaic expressions, especially of routine formulae. It should not be forgotten that conversational interaction is of a multimodal nature, comprising phonetic, prosodic, morphosyntactic, semantic, pragmatic (social, contextual, sequential, stylistic...), and, last but not least, non-verbal components – all of them play or can play a role in the production and interpretation of formulaic sequences.

Possibly due to extensive treatment of widely researched “substantive idioms” or phraseological expressions, even though the analysis of their pragmatic functions has so far not received sufficient attention, Fillmore, Kay & O’Connor (1988) focus their research on the second type of “idiomatic constructions” mentioned:

Formal idioms, […], are syntactic patterns dedicated to semantic and pragmatic purposes not knowable from their form alone. It is the formal idioms which raise the most serious theoretical issues, and which hold our main interest in this paper. (Fillmore, Kay & O’Connor 1988: 505-506)

Even if these “formal idioms” cannot be generated via the rules of an “atomistic grammar”, they nevertheless possess a certain type of grammatical structure:

Such expressions have grammatical structure, […], but the structures they have are not made intelligible by knowledge of the familiar rules of the grammar and how those rules are most generally applied. (Fillmore, Kay & O’Connor 1988: 505)

But it is not only from a theoretical perspective that multifactorial formal constructions are particularly interesting for research on language acquisition, they are also highly relevant for efficient language learning because– due to their lexical flexibility – they offer a by far larger productive potential than lexically filled “substantive idioms” which are in any case hardly suitable for non-natives (cf. supra). As a consequence, the present paper will concentrate on “formal idioms” or “(formal) constructions”. However, it seems difficult to differentiate idiomatic and non idiomatic constructions on the basis of the definition of idiomaticity quoted earlier in this section, all the more as someone who

merely knew the grammar and the vocabulary of the language could not, by virtue of that knowledge alone, know (i) how to say it, or (ii) what it means, or (iii) whether it is a conventional thing to say (Fillmore, Kay & O’Connor 1988: 504).

In fact, even though a (non-native) speaker may be able to understand a semantically transparent construction (cf. the 2nd criterion), e.g. of the “the x-er, the y-er” type, he need not necessarily know “how to say it” (cf. the 1st criterion), and especially not “whether it is a conventional thing to say” (cf. the 3rd criterion). From this point of view, even a priori semantically non idiomatic constructions like “The more carefully you do your work, the easier it will get” (Fillmore, Kay & O’Connor 1988: 506) of the “The x-er, the y-er” type would have to be considered as “idiomatic constructions” in a wider sense. In any case, idiomaticity and compositionality of constructions subject to multiple constraints seem to be a question of degree, not, however, an “existential” question.

Before attempting to exemplify this wider understanding of “formal idioms / constructions” in chap. 4, different examples of constructions in the German language, so far discussed, will be briefly presented in sub-section 3.2.

3.2     Different Types of Constructions in German

For the time being, very few formal constructions which correspond to the acceptation defined have been specifically described for FLL. One of the rare treatises is Handwerker &  Madlener’s (2009) “Chunks für DaF” dealing with “sein + participle I or II” of a verb designating a psychic disposition or emotional state[27] such as begeistern or langweilen, i.e. begeistert sein or gelangweilt sein.[28] However, this approach, following Goldberg’s (2003) very wide understanding of constructions[29], appears to be too unspecific in order to be operational for foreign language methodology, the more so as multifactorial application criteria can hardly be described for such a narrow type of construction. 

On the other hand, Goldberg describes “caused-motion-constructions” of the “x causes y to move towards z” type, which can be non idiomatic, e.g. He puts the bottle in the fridge, or idiomatic, e.g. He drank himself to his grave. However, whereas the non-idiomatic type of construction in a very wide sense might be suitable for FLL, the second idiomatic type is definitely not suitable for reasons developed earlier in this paper. The same argument holds true for other types of constructions which have been discussed of late:
  • the “let alone” construction analyzed by Fillmore, Kay & O’Connor (1988), which is by far too idiomatic and complicated for a foreign language learner;
  • many of the “discourse markers” of German described by Auer & Günthner (2003), constituting deviations from the standard – written – grammar of German, such as obwohl, weil, wenn, wo, wobei, serving in “main clause” constructions with the finite verb in second and not in final position; an extremely high degree of proficiency in the foreign language would be necessary in order to use this type of construction;
  • Barden, Elstermann & Fiehler’s (2001) „operator-scopus-constructions“, two-part construction units, consisting of a first part such as however, to cut a long story short, let’s say or mind you[30], which provide a “guide” for the interpretation of a second part consisting of a complete structure, e.g. He’s a very honest person, mind you, he never told me that he won the pools. As the former types mentioned, the use of this construction necessitates a near-native degree of competence;
  • the so-called “absentive”-constructions (König 2009) – „subject + finite form of sein + infinitive of verb designating an action” – such as [Sie ist im Moment nicht da,] sie ist einkaufen or Er ist arbeiten [und kommt erst heute Abend zurück][31] are equally reserved for learners with a high level of proficiency .

In the face of this statement as to the difficulty of determining FLL-adequate formal idioms, how can constructions for FLL be determined? Even if the investigation of vast corpora of conversational material were possible, Wray & Namba (2003) state one important reason why such a method would not be appropriate:

[…] there is as yet no means of categorically isolating formulaic sequences in a consistent way across data types. The main reason for this difficulty is that the majority of formulaic sequences are, to the casual ear or eye, indistinguishable from novel strings because they are grammatically unexceptional and their meaning is entirely predictable. (Wray & Namba 2003: 26)

This is why we should like to propose a corpus-based approach[32], using as a starting point existing listings of sentence models or notions / functions, e.g. 23 primary and 18 secondary sentence models of the Duden Grammatik (Drosdowski 1984) or of the notions presented in Threshold 1990 (van Ek & Trim 1990)[33]. Forms and functions of the aforementioned models or notions will be examined within large corpora of spoken and written language in order to deliver a multifactorial description – including morphosyntactic, semantic, prosodic, contextual, nonverbal elements – of FLL-relevant constructions.

Before making an attempt in section 5 towards the description of such an FLL-oriented construction, in the following passages we will deal with the description of the difficulties which learners of German encounter with one specific grammatical phenomenon.

4    FLL Difficulties with the German Verbs müssen and sollen

In order to be able to express / deny or to enquire about degrees of probability or necessity (cf. the notions in Threshold 1990, van Ek & Trim 1991), learners of German have to master the modal and modalizing verbs müssen and sollen. However, for French speakers, these are particularly difficult to use in an appropriate manner, even for those who otherwise demonstrate a considerably advanced level of proficiency. Some extremely unlikely answers found in a 25-item test carried out among students both at school and at university in France demonstrate the complexity of the use and the differentiation of the two verbs.

(a) Du, ich *soll dir unbedingt was sagen: ich habe einen neuen Computer.
(b) *Sollst du eigentlich immer so laut schreien?
(c) Hausarbeiten *muss man nicht ins Uferlose ausdehnen.
(d) *Muss ich dir irgendwas aus der Stadt mitbringen?
(e) Du *sollst wirklich erst überlegen, bevor du so einen Quatsch erzählst.
(f) Wann *sollst du normalerweise morgens aufstehen? Meistens schon um 6.
(g) Wo *muss ich das jetzt so schnell herbekommen?

The preceding answers are considered as unusual or even erroneous as they would be possible exclusively in highly specific and peculiar contexts, e.g. in item (a), the use of sollen implies the following: I was told to tell you that I’ve got a new computer. In theory, this is imaginable, but it is highly unlikely in reality. What is more, the test items were not invented, but based on the detailed analysis of conversational utterances from authentic contexts within two conversational corpora: our own corpus of German telephone conversations[34] and from the corpus Sprechstundengespräche (Boettcher et al. 2005). The expected – adequate – answers are, thus, not all arbitrarily imposed, but they are the result of authentic use in authentic – non exceptional – communicative situations.

Based on this collection of recurrent müssen-sollen utterances, six basic rules for the use of these modal or modalizing verbs were defined (cf. infra). These rules were then presented to one of the test groups having produced rather mediocre results. Without discussing and “correcting” the different test items, but illustrating the rules via the same types of use as in the questionnaire, the same cloze test was subsequently submitted to the same 27 students from a first-year university course of German as a foreign language. The test results were as disappointing and surprisingly inadequate as the first time.

The following arguments can be proposed to explain this failure:
  • The rules and explanations provided to the students of the test group were obviously insufficient. But as these rules were specifically developed for the test items and thoroughly explained and discussed with the learners, there seem to be more profound reasons.
  • Adult learners in their eighth year of German may have developed erroneous – bad – habits in the use of müssen / sollen which have not been eradicated by the “new” explanations provided.
  • The “bad” habit may consist in translating – erroneously – from French devoir, which unfortunately is of not much help as Je dois aller en ville can be translated both by German Ich muss in die Stadt (I have to go into town) or Ich soll in die Stadt (I am supposed to go into town).
  • Finally, even at an advanced learner’s stage, the cumulated presentation of the complete rules for the production of müssen / sollen utterances may create more confusion than help resolve existing problems with an adequate use of these modal or modalizing verbs.

In order to counter the prevalent confusion of müssen and sollen, the following section of this paper proposes the development of basic müssen / sollen constructions as well as reflexions on their learning and teaching.

5      Müssen / sollen Constructions in German and Principles for their Learning and Teaching

Relevant constructions for FLL obviously have to be elaborated on the basis of sufficiently large and representative corpora. Language models cannot be devised on the basis of the sole intuition of their authors[35]. In fact, corpus-based research alone can guarantee the development of realistic and learner-adequate models or constructions. For müssen / sollen, following the study of the two corpora mentioned, seven basic constructions accomplishing the main functions of the two verbs can be retained for FLL:

1)  müss- expressing an obligation or necessity the grammatical subject cannot avoid because of the circumstances or laws of nature. However, this obligation is in no case explicitly imposed by another person or instance; the obligation can also be denied or a question can be asked about it: “subject + müss- + infinitive (phrase)“, e.g. Ich muss einkaufen, Ich muss Hausaufgaben machen, or „subject + müss- + prepositional phrase“, e.g. Ich muss in die Stadt, Wann musst du abends ins Bett?

2)   soll- expressing – denying or asking for – an obligation or request evidently formulated by a third party or moral / legal instance and which is imposed or suggested to the grammatical subject: “[appellative] + [indication of “obligator”] + subject + soll- + infinitive phrase”, e.g. Peter, deine Mutter hat gesagt, du sollst sofort nach Hause kommen or the question format: “soll + subject + infinitive phrase”, e.g. Soll ich dich am Bahnhof abholen?

3)   sollt- in a subjunctive II affirmative or negative form expressing advice or asking for it: “subject + sollt- + infinitive phrase”, e.g. Du solltest regelmäßig deine Vokabeln lernen, Du solltest nicht so viel trinken, or “sollt- + subject + infinitive phase”, e.g. Sollte ich bessser das grüne oder das rote Kleid anziehen?

4)  modalizing “hearsay soll-“ in order to express a fact the speaker has learned from a third party, the substitution of a modalizer such as angeblich, anscheinend or of an introductory phrase like Ich habe gehört, ... can function as a test. The basic modalizing soll- structure would be: “subject + soll- + infinitive phrase”, e.g. Er soll krank sein (= Er ist angeblich krank), Sie soll aus München kommen (= Ich habe gehört, sie kommt aus München).

5)   modalizing müss- expressing a high degree of certainty or a strong assumption about the veracity of the fact enunciated. As a test a substitution by a modalizer of the wahrscheinlich, mit ziemlicher Sicherheit type can be applied. A basic müss- modalizing construction could be: “subject + müss- + noun phrase OR adjective (phrase) + sein”, e.g. Er muss krank sein, [sonst wäre er bestimmt gekommen] (= Er ist wahrscheinlich krank).

6)  sollt- in a subjunctive II form expressing the hypothetical nature or the irreality of the fact enunciated within a conditional format which could be replaced by a wenn / falls-construction. A basic sollt- conditional structure could be: “sollt- + subject + infinitive phrase + finite verb + subject + infinitive phrase”, e.g. Solltest du Zeit haben, könntest du mir einen Gefallen tun  = Falls du Zeit hast, könntest du mir einen Gefallen tun.[36]

7)   Various müss-/soll-constructions in between formal and substantive idioms, i.e. which are more or less lexically “filled”. Even if these genuinely idiomatic constructions in a narrower sense than the aforementioned preserve some of the original semantics of müss-/soll-, they express by no means obligation or necessity like in (1) or )2). Some examples of a certainly much bigger class which would necessitate extensive corpus research are:
    • was soll + noun phrase?”, e.g. Was soll das / der Quatsch / der Mist etc.?, expressing discontentment.
    • muss das sein, [dass + VP]?”, e.g. Muss das sein? Muss das sein, dass du ständig zu spät kommst?, indicating disagreement or dissatisfaction with respect to a third party’s preceding action.
    • das + hättest + subject + infinitive + sollen”, e.g. Das hättest du nicht tun / sagen / essen sollen, expressing disapproval of a preceding act.
    • was/wie/wo + soll- + subject + [complement] + infinitive?”, e.g. Was soll ich dazu sagen?, Wie soll ich das [jetzt so schnell] finden?, Wo soll ich das [jetzt] herbekommen? as a rhetorical question indicating that the speaker does not want to or cannot provide an answer or execute an implied action.
    • “interrogative pronoun + soll- + subject + [complement] + infinitive phrase”, e.g., Wem soll ich das geben?, Wann soll ich kommen?, Warum soll ich das machen?, genuinely asking for information. However, the aforementioned pronouns was / wie / wo can both be used for rhetorical and genuine questions.
It goes without saying that the aforementioned seven müss- / soll-constructions have to be thoroughly differentiated as to learners’ proficiency levels. In fact, except for constructions (1) and (2), the remaining models (3) to (7) are strictly reserved for learners with a (highly) developed competence as semantic, contextual, social conditions of use seem a priorifar too complex for the average learner.

Apart from the systematic corpus-based development of constructions corresponding to learners’ specific communicative needs, the teaching and learning of these should invariably be based on the following methodological principles:
  • Constructions should be presented in authentic, learner-relevant contexts and preferably be discovered inductively. This implies that the teacher does not provide highly theoretical explanations of the different forms and functions of the construc­tions. It would, for instance, be superfluous to tell a learner that the constructions (3) or (6) contain a subjunctive II form. All he needs to know is that he has to use sollt- if he wants to produce a piece of advice. Obviously, theoretical elements can be part of a cognitive approach. However, this is no excuse for providing lengthy and unnecessary theoretical explanations which would create confusion rather than enhanced communicative performance.
  • Constructions should systematically serve the accomplishment of specific learner-relevant conversational activities corresponding to their proficiency level.
  • In order to avoid confusion of two or several constructions, one and only one construction at a time should be introduced and practiced until it is definitely mastered by learners. Only then should a new construction be introduced.
  • Translation from French or other languages into German has to be proscribed.

In principle, it seems possible, but also preferable, to systematically describe any language phenomenon in terms of constructions which can be acquired as a whole by foreign language learners. Starting out from grammatical phenomena which are particularly difficult to master for French learners of German, as our experience in foreign language teaching tells us, we intend to describe in the same way constructions for the use of the German passive voice, differentiating between dynamic and stative passive forms; contracted vs. non-contracted forms of noun phrases; identification vs. non-identification of noun phrases or the place of the negator. And, as a particular challenge, we should like to describe German declension – more and more difficult for native and non-native speakers – on the basis of a thorough corpus research as recurrent and typical constructions rather than as abstract and context-free rules which obviously do not produce the expected results.


Aguado, Karin (2002). Formelhafte Sequenzen und ihre Funktionen für den L2-Erwerb. Zeitschrift für Angewandte Linguistik 37, 27-49.

Altenberg, Bengt (1998). On the Phraseology of Spoken English: the Evidence of Recurrent Word Combinations. In Cowie, Anthony Paul (ed.) (2004). Phraseology: Theory, Analysis and Applications. Oxford: OUP, 101-122.

Auer, Peter / Günthner, Susanne (2003). Die Entstehung von Diskursmarkern im Deutschen – ein Fall von Grammatikalisierung? InLiSt 38, 1-30.

Bally, Charles (1909). Traité de Stylistique Française. Paris: Klincksieck.

Barden, Birgit / Elstermann, Mechthild / Fiehler, Reinhard (2001). Operator-Skopus-Strukturen in gesprochener Sprache. In Liedtke, Frank / Hundsnurscher, Franz (eds.). Pragmatische Syntax. Tübingen: Niemeyer, 197-233.

Bloomfield, Leonard (1933). Language. London: Allen and Unwin.

Boettcher, Wolfgang / Limburg, Anika / Meer, Dorothee / Zegers, Vera (2005). Sprechstundengespräche an der Hochschule. Ein Transkriptband. Radolfzell: Verlag für Gesprächsforschung.

Bolinger, Dwight (1976). Meaning and Memory. Forum Linguisticum 1, 1-14.

Bréal, Michel (18797/1982). Essai de sémantique (science des significations). Paris: Hachette/Brionne: Montfort.

Burger, Harald (unter Mitarbeit von H. Jaksche) (1973). Idiomatik des Deutschen (= Germanistische Arbeitshefte, 16). Tübingen: Niemeyer.

Burger, Harald (20104, 1998). Phraseologie. Eine Einführung am Beispiel des Deutschen. Berlin: Erich Schmidt.

Burger, Harald / Buhofer, Annelies / Sialm, Ambros (1982). Handbuch der Phraseologie. Berlin/New York: de Gruyter.

Chafe, Wallace (1968). Idiomaticity as an Anomaly in the Chomskyan Paradigm. Foundations of Language 4/2, 109-127.

Coulmas, Florian (1981). Routine im Gespräch. Zur pragmatischen Fundierung der Idiomatik. Wiesbaden: Athenaion.

Coseriu, E. (1967). Lexikalische Solidaritäten. Poetica 1/3, 293-303.

Council of Europe (2001). Common European Framework of Reference for Languages: Learning, Teaching, Assessment. Strasbourg: Council of Europe. URL: http://www.coe.int/t/dg4/linguistic/Source/Framework_EN.pdf (20/11/2011)

Dobrovol'skij, Dmitrij / Lûbimova, Natalia (1993). ‚Wie man so schön sagt, kommt das gar nicht in die Tüte’ – Zur metakommunikativen Umrahmung von Idiomen. Deutsch als Fremdsprache 30, 151-156.

Donalies, Elke (2009). Basiswissen Deutsche Phraseologie. Tübingen / Basel: A. Francke.

Drosdowski, Günter et al. (eds.) (1984). Duden 4 – Grammatik der deutschen Gegenwartssprache. 4. völlig neu bearbeitete u. erweiterte Auflage. Mannheim / Wien / Zürich: Dudenverlag.

Erman, Britt / Warren, Beatrice (2000). The Idiom Principle and the Open Choice Principle. Text 20/1, 29-62.

Feilke, Helmuth (1998). Idiomatische Prägung. In Barz, Irmhild / Öhlschläger, Günther (eds.). Zwischen Grammatik und Lexikon. Tübingen: Niemeyer, 69-80.

Fleischer, Wolfgang (1982, 1997 (2. durchges. u. erg. Aufl.)). Phraseologie der deutschen Gegenwartssprache. Tübingen: Niemeyer.

Fillmore, Charles J. / Kay, Paul / O’Connor, Mary Catherine (1988). Regularity and Idiomaticity in Grammatical Constructions: The Case of Let Alone. Language 64/3, 501-538.

Firth, John R. (1937, 1964). The Tongues of Men and Speech. Oxford: OUP.

Gasparov, Boris (2004). Speech, Memory, and Meaning: Intertextuality in Everyday Language. Language and Cognition 2003-2004. University Seminar. Columbia University. New York. URL: http://www.columbia.edu/~remez/03-04compiled.pdf (26/02/2010)

Goldberg, Adele E. (2003). Constructions: A New Theoretical Approach to Language. Trends in Cognitive Sciences 7/5, 219-224.

Grunig, Blanche-Noëlle (1997). La locution comme défi aux théories linguistiques. In Martins-Baltar, Michel (ed.). La locution, entre langue et usages. Paris: ENS/Ophrys, 225-240.

Gülich, Elisabeth (2008). Le recours au préformé: une ressource dans l’interaction conversationnelle. In Durand, Jacques / Habert, Benoît / Laks, Bernard (resp.) (2008). Congrès mondial de linguistique française. Paris, 9-12 juillet 2008. Recueil des résumés + contribution sur CDRom. Paris: ILF/EDP Sciences, p. 93. Disponible en ligne sous: http://www.linguistiquefrancaise.org/index.php?option=article&access=doi&doi=10.1051/cmlf08315 (16/11/2009)

Hallsteinsdottir, Erla (2011). Aktuelle Forschungsfragen der deutschsprachigen Phraseodidaktik. Linguistik online 47/3, 19.

Handwerker, Brigitte & Madlener, Karin (2009). Chunks für DaF. Theoretischer Hintergrund und Prototyp einer multimedialen Lernumgebung (inklusive DVD). Baltmannsweiler: Schneider Verlag Hohengehren.

Hoey, Michael (2005). Lexical Priming. A New Theory of Words and Language. New York: Routledge.

Jesenšek, Vida (2006). Phraseologie und Fremdsprachenlernen. Zur Problematik einer angemessenen phraseodidaktischen Umsetzung. Linguistik online 27/2, 137-147.

Jespersen, Otto (1924/1971). La philosophie de la grammaire. Traduit par A. M. Léonard. Paris: Minuit.

Kay, Paul / Fillmore, Charles (1999). Grammatical Constructions and Linguistic Generalization: the What’s X doing Y? Construction. Language 75/1, 1-33. [Cité d’après le preprint (1997) plus disponible sur le site de Paul Kay; 40.]

König, Svenja (2009). ‚Alle sind Deutschland … außer Fritz Eckenga – der ist einkaufen!‘ Der Absentiv in der deutschen Gegenwartssprache. In Winkler, Edeltraud (ed.). Konstruktionelle Varianz bei Verben. Opal-Sonderheft 4, 42-74.

Lorenz-Bourjot, Martine / Lüger, Heinz-Helmut (eds.) (2001). Phraseologie und Phraseodidaktik. Wien: Praesens.

Lüger, Heinz-Helmut (2007). Michel Bréal – ein bedeutender Sprachwissenschaftler? In Giessen, Hans / Lüger, Heinz-Helmut / Volz, Günter (eds.). Michel Bréal – Grenzüberschreitende Signaturen. Landau: VEP.

Martin, Robert (1997). Sur les facteurs du figement lexical. In Martins-Baltar, Michel (ed.). La locution, entre langue et usages. Paris: ENS/Ophrys, 291-305.

Mejri, Salah (2007). French phraseology. In Burger, Harald / Dobrovol’skij, Dmitrij / Kühn, Peter / Norrick, Neil R. (eds.). Phraseologie/Phraseology. Ein internationales Handbuch der zeitgenössischen Forschung / An International Handbook of Contemporary Research. Berlin / New York: de Gruyter, 682-691.

Mel’čuk, Igor (1998). Collocations and Lexical Functions. In: Cowie, Andrew P. (ed.). Phraseology. Theory, Analyses, and Applications. Oxford: OUP, 23-53.

Paul, Hermann (1880/101995). Prinzipien der Sprachgeschichte. Tübingen: Niemeyer.

Pinker, Steven (1994). The Language Instinct: How the Mind Creates Language. New York: William Morrow.

Porzig, Walter (1934). Wesenhafte Bedeutungsbeziehungen. Beiträge zur Geschichte und Sprache der deutschen Literatur 58, 70-97.

Sapir, Edward (1921). Language. New York: Harcourt, Brace, Jovanovitch.

Saussure, Ferdinand de (1916, 1995). Cours de linguistique générale. Paris: Payot.

Schmale, Günter (1984). excuse can you tell me the way to George Square – Die linguistische Beschreibung von Interaktionseinheiten für den Fremdsprachenunterricht am Beispiel von Asking/Telling Someone the Way. Bielefelder Beiträge zur Sprachlehrforschung 1, 26-51.

Schmale, Günter (2009). Phraseologische Ausdrücke als Bestandteil des Fremdsprachenerwerbs – Überlegungen zur Phraseodidaktik auf der Grundlage einer korpusbasierten Analyse deutscher Talkshows. In: Bachmann-Stein, Andrea / Stein, Stefan: Mediale Varietäten – Analysen von gesprochener und geschriebener Sprache und ihre fremdsprachendidaktischen Potenziale. Beiträge zur Fremdsprachenvermittlung. Sonderheft 13, 149-179.

Schmale, Günter (2010). Ist ein idiomatischer Ausdruck immer expressiv? Korpusbasierte und fragebogengestützte Beobachtungen zu einer verbreiteten Prämisse. In Kuiper, K. (ed.). Yearbook of Phraseology. Berlin / New York: de Gruyter Mouton, 97-124.

Schmale, Günter (2011). Was ist in der Sprache ‚vorgeformt‘? Überlegungen zu einer erweiterten Definition sprachlicher Präformiertheit. In Schäfer, Patrick / Schowalter, Christine (eds.). In mediam linguam. Mediensprache – Redewendungen – Sprachvermittlung. Festschrift für Heinz-Helmut Lüger zum 65. Geburtstag. Landau: VEP, 177-190.

Schmale, Günter (2012). Morpho-Syntax oder präformierte Konstruktionseinheiten – Welcher linguistische Ansatz für das Fremdsprachenlernen? In DAAD (ed.): Zukunftsfragen der Germanistik. Beiträge der DAAD-Germanistentagung 2011 mit den Partnerländern Frankreich, Belgien, Niederlande, Luxemburg (= Germanistik im Dialog). Göttingen: Wallstein, 195-209.

Sinclair, John (1991). Corpus, Concordance, Collocation. Oxford: OUP.

Tognini-Bonelli, Elena (2001). Corpus Linguistics at Work. Amsterdam: Benjamins.

Van Ek, Jan A. / Trim, John L.M. (1998, 1991). Threshold 1990. Council of Europe: CUP.

Van Lancker-Sidtis / Diana / Rallon, Gail (2004). Tracking the incidence of formulaic expressions in everyday speech: methods for classification and verification. Language & Communication 24/3, 207-240.

Widdowson, Henry G. (1989). Knowledge of language and ability for use. IApplied Linguistics 10, 128-137.

Wray, Alison (2000). Formulaic Sequences in Second Language Teaching: Principle and Practice. IApplied Linguistics 21/4, 463-489.

Wray, Alison / Namba, Kazuhiko (2003). Use of Formulaic Language by a Japanese-English Bilingual Child: A Practical Approach to Data Analysis. Japan Journal of Multilingualism and Multiculturalism. 9 / 1, 24-51.

Wray, Alison / Perkins, Michael R. (2000). The functions of formulaic language. An integrated model. Language & Communication 20, 1-28.

[1] Cf. also Paul (1880): „Erst wo Sprechen und Verstehen auf Reproduktion beruht, ist Sprache da.“ „Only where speaking and understanding are based on reproduction, language will come into existence“ (translation GS)
[2]  Quoted following Lüger (2007: page number?).
[3]  Keeping in mind that these treatises do not always share Bolinger’s positive understanding of language formulaicity, see, for instance, Saussure or Bloomfield.
[4]  Our translation of Feilke’s term “Idiomatische Prägung”.
[5]  Which, of course, is the exact opposite of Pinker’s (1994) credo: “Virtually every sentence that a person utters or understands is a brand new combination of words, appearing for the first time in the history of the universe.” (quoted from Gasparov 2004: 45), to our mind, counterintuitive at first sight.
[6]  On the other hand, given the fact that phraseological research has known a considerable extension of its field of study in recent years (Schmale 2011), one might argue that even more elements than suspected so far might be prefabricated in language.
[7]  An assertion fiercely rejected by Chomskyan generativists: „[…], the Chomskyan tradition has […] tenaciously challenged the idea that it (i.e. formulaicity; GS) plays anything more than a peripheral role in language production and comprehension.“ (Wray & Perkins 2000: 9)
[8]  Translated from the German “Phraseodidaktik”; cf. e.g. Lorenz-Bourjot & Lüger (2001), Schmale (2009).
[9]  But also in a number of countries in Eastern Europe (e.g. Jesenšek 2006).
[10] Burger’s “structural phrasemes”, serving to establish syntactic relations, e.g. neither – nor, will be dealt with later in this paper, under the heading of “constructions”.
[11] As opposed to idiomatic expressions whose semantic structure is non compositional (cf. infra).
[12] German: sich die Zähne putzen; French: se laver les dents.
[13] If he were blind, one would have to say ein wirklich blinder Passagier or ein blinder blinder Passagier in German.
[14] “Phraseologische Termini” in German (Burger 20104: 49).
[15] German: Es ist nicht alles Gold, was glänzt; French: Tout ce qui brille n’est pas or.
[16] English: Out of sight, out of mind; French: Loin des yeux, loin du coeur.
[17] The latter being partially idiomatic, drunk having “normal” semantics whereas as a lord would have a non-compositional meaning equivalent to extremely or immensely.
[18] Admittedly, however, they are probably less expressive (Schmale 2010).
[19] This is an observation which perfectly corresponds to our own experiences in English or French exolingual situations as a native speaker of German.
[20]   The same comment applies to playing on words with phrasemes, which is scarcely tolerated when realised by non-native speakers.
[21] The essence of which is, by the way, by no means new; cf. for instance Chafe (1968).
[22] However, also see Bolinger (1976), Hoey’s (2005) concept of “lexical priming” or Sinclair (1991: 65): “It seems that there is a strong tendency for sense and syntax to be associated.”
[23] Knowing that formal idioms can serve as host to substantive idioms” (id.: 506), e.g. The bigger they come, the harder they fall using the “the x-er, the y-er” structure.
[24] Or rather lexically more or less “open” as “formal idioms” or non-idiomatic constructions also quite regularly contain lexical material; cf. the aforementioned “the x-er, the y-er” structure.
[25] “Die Konstruktionen liegen in einem Grenzbereich der Phraseologie zur Syntax. Ihre Einbeziehung in die Phraseologie ist strittig.“ (Fleischer 1982: 135)
[26] Cf. also Grunig (1997), who uses the term “schéma”, or Martin (1997), who employs the term “modèle locutionnel”.
[27] The authors use the term “psychische Wirkungsverben” in German.
[28] The English translation of these verbs is to be enthusiastic or bored.
[29] She even considers morphemes, words or complex words as constructions (Goldberg 2003: 220).
[30] The authors mentioned discuss, for instance, kurz und gut, sicher, allerdings, obwohl..
[31] The respective English translations are: She’s not in at the moment, she’s gone shopping. – He’s gone to work and will be back tonight.
[32] Which would, in Tognini-Bonelli’s (2001) terms, not be, strictly speaking, “corpus-driven”.
[33] A thoroughly revised and extended edition of The Threshold Level in a European unit / credit system for modern language learning by adults (1975).
[34] Used for our doctoral thesis (Schmale 1995).
[35] Cf. Schmale (1984), who demonstrates that textbook dialogue models such as Asking / Telling Some­body the Way do not correspond to what the researcher encounters in recorded and transcribed authentic situations of this type.
[36] Which could function as a declarative or an interrogative utterance.