Especially in spoken language, phenomena like elision or fusion occur fairly frequently, and often to an extreme degree. Take the German sentence 'Das ist ein Besen.', which I might (though not always) produce as /dasnbeːzn/.

Unless in a highly prescriptivist context, these are generally not seen as errors or mistakes, and in any case are often predictable and thus not resulting from any permanent or temporary confusion about applicable rules. Instead, elision, fusion and the like seem to be occurring in these cases mainly for ease and speed.

In googling this question, the concept of "ease of articulation" came up, but this seems to mainly refer to the actual ease of articulation and resulting change of single sounds, and less to the cross-segmental phenomenon I'm aiming at.

  • The economy of effort aka laziness is behind it all, behind all the progress of our civilization, not only behind language development.
    – Yellow Sky
    Jul 16 at 16:50
  • 2
    The technical term you are looking for is: connected speech. Here is a non-technical introduction to it: teachingenglish.org.uk/article/connected-speech [It exists in all languages]
    – Lambie
    Jul 16 at 20:31

2 Answers 2


"Ease of articulation" is in fact the closest most-widely used term that covers the phenomena that you listed, however that motivation is often invoked improperly. Supposed "ease of articulation" is often invoked when in fact the motivation for a phonetic change is a perceptual factor (making the sounds more audible, or eliminating the sound because it is inaudible).

For the specific example, we can sub-categorize the motivation with reference to timing of articulatory movements and their acoustic consequences, and note that in the output, gestures overlap so that the vocalic gestures are obscured by the [s]-gestures, and you don't hear them. One would have to undertake an articulatory study to determine whether the vowel gestures are simply muted in acoustic value, or are they actually physically omitted.

This explanation does depend on there being certain precursors which are not about gestural overlap, namely that the magnitude of the vowel gestures are independently reduced for some reason. For your example, the precursor motivation is about speech rhythm – the explanation for why you don't get such reduction in "Daß ißt ein Besen, nicht zwei" (hard to come up with a more plausible contrasting example).

  • One recurrent fact is that predictable chunks at the beginning or end of a word can drop off in normal speech because they're predictable. The end of a word is silence, and that can wash inward, modifying or deleting sounds that don't seem needed. That's what happened to Latin, destroying the case suffix paradigms by a set of normal phonological changes like not distinguishing final long and short vowels and deleting final /s/. Now we have Romance languages with no cases instead.
    – jlawler
    Jul 16 at 15:01

I suggest that the phonetic causes of sound change are not entirely understood, and mistaken assumptions about normative language is responsible. This is critical as a matter of attitude towards language, because it may influence our choice of words for the sought description.

Your "highly prescriptive context" has a low threshold. The premise is doubtful, namely that "phenomena like elision or fusion occur fairly frequently" as deviating from the norm. The contrast is certainly notable, but it requires a far broader context to interpret.

Strictly speaking, the premise is always difficult to establish since spoken registers are unlikely recorded in historical writing, or difficult to decipher.

"Phonetic models of sound change differ as to whether the conversion of a phonetic bias into a permanent, obligatory feature of a language happens primarily for perceptual or articulatory reasons" [Stevens, M. & Harrington, J., (2022) “Individual variation and the coarticulatory path to sound change: agent-based modeling of /str/ in English and Italian”, Glossa: a journal of general linguistics 7(1). doi: https://doi.org/10.16995/glossa.8869]

NB: *Claire Bowern argues about exceptions of mechanisms of regular sound change, "... that the standard explanations are inadequate, and possible solutions have not yet been empirically investigated." (Bowern, C. (2022) Where have all the sound changes gone? Phonological stability and mechanisms of sound change. Linguistics Vanguard, (). https://doi.org/10.1515/lingvan-2021-0073)

This also holds for a pluricentric language like German, which is on one hand a cover term for a number of varieties which may be recognized as own languages, and an idealized lingua franca for international communication on the other hand. The distinction is difficult to maintain.

Moreover, perception has been problematicized as an issue of discrimination because stereotyping and prejudice can lead to oppression. Notable examples pertain to the Fair Housing Act and the Civil Rights Act 1968 as amended:

[HUD vs. Ross (1994)] and a few others (e.g., City of Chicago v. Matchmaker Real Estate Sales Center, 1992) with respect to Ms. Frazier’s “accent”) demonstrate that individuals are capable of being held liable because of their auditory assessment of a speaker.

However, not all jurists agree that racial identity is ascertained in the absence of visual prompting.

[Purnell, T. Idsardi, W. Baugh, J. (1999) Perceptual and phonetic experiments on American English dialect identification. Journal of language and social psychology 18 (1), 10-30]

Although attitude towards language is under research and methods are developed, (eg. 2012) This is being studied and chiefly rejected in linguistics.

"As a language scientist and one of the foremost fathers of variationist sociolinguistics, Labov has led the way in developing theory and providing legitimacy to what he refers to as African American Vernacular English (AAVE), a language variety that was once thought to be sloppy speech, errors, and a sign of mental retardation. Labov's (1972a) work has argued the case of “Academic Ignorance and Black Intelligence,” wherein he has showed that Black speakers display intricate and complex narration, reasoning, and argumentative skills. He demonstrated the systematicity of Black vernacular speech, in terms of its grammaticality, syntax, morphology, and phonology." [When Language Is and Not the Issue: The Case of “AAVE” Literacy Research, Teaching, and Labov's Prescription for Social (in) Equality Elaine Richardson The Routledge handbook of educational linguistics, 185-194, 2014]

Indeed it has been said that AAVE retains features that are obsolete in standard American English. Many dialects do, as much as they innovate.

To answer the question about "Ease of Articulation" or connected speech, one had to be aware of the context. That said, those egreggious cases differ from the given example in their historic development. This will be left for fuether research into Middle High German and Kanzleisprache through Luther upto modern Jugendsprache (eg. "I bims", cp. Old High German bim, Old Saxon beum, Old English beom 1PSg. "am"; "Lass ma McDonalds gehen", "Bra", etc.).

In the given case, /dasnbeːzn/ may instead be understood as synthetic construction. It might sound far fetched to etymologize this differently than das ist ein Besen, but hear me out. Da is a deictic particle, whereas das may be dat or even dit (with a ultra short vowel). -s is often enclitic. The Germanic parent language had probably no definite articles, as Norse indicates: definite nouns are suffixed, eg. -en. The verbal auxilary paradigm "to be" also underwent sweeping changes, with sein now as infinitive related to indicative 3PSg. indicative is and conjunctive sei (PIE *h₁és- ~ *h₁s- [Ringe, Taylor. (2006) From Indo-European to Old English]). Add to that, the origin of n-stems is uncertain as far as I know (NB: Guus Kronen has worked on that, reinstating Kluge's Law, but it's not widely accepted and I haven't understood the theory, still).

Consequently, we might try to read dasenbesne variously, emendate das [hier] ist, or. The indefinite 'n may be hypercorrect, nasalized, retained from the nasal in 3PSg. *h₁sénti if plural (see below), whatever.

It is not unreasonable to believe that Besen is a collective, of straw as it were. It is comparable with Samen "seed", (obsolete) Same, which shows historic vexilation in numerus and maybe genus. Proto-West-Germanic *besmō suggests an o-stem. Same as for Samen, the m possibly reflects *-m̥n: wiktionary offers two competing opinions on the PIE derivation, eitherway deverbal (en.WT). The semantics of *-m̥n may be decisive at this point, but the more interesting fact is that the -n appears only around the time that Low German steps up. This -n is probably explained in the literature as analogy.

Low German is historically connected to Skandinavia and the Balticum through the Hanse (itself derived from *ḱom- [en.WT]).

NB: Wolfgang Pfeifer argues without Norse comparanda for "westgerm. *besman- ‘Rute, Besen’.", assuming loans into Norse, apparently, and "Da jedoch nur westgerm. Formen bezeugt sind, kann Herkunft des Wortes auch aus einer nicht-ie. Sprache eines vorgerm. Substrats angenommen werden." [„Besen“, bereitgestellt durch das Digitale Wörterbuch der deutschen Sprache, https://www.dwds.de/wb/Besen, abgerufen am 18.07.2022.]

Other comparisons come to min: Pinsel "brush", cp. Latin penicilum; or Bürste "brush", see Engl. burst > bust etc. for reference, note Besenstiel "broomstick" (cf. DWB, alas not dated by Pfeifer; cp. *-dhlom > Pgem. *-īlaz, cf. sta-bulum) and perhaps unrelated Besenreißer "spider veins" (cp. bruise, cf. en.WT, especially with reference to Old Church Slavonic бръснути (brŭsnuti), "to rake"), to bash, bush, bushel.

Long story short, it's entirely uncertain what this is.

Finally, if it could be anything, cp. Dasein "existence".

NB: Olav Hackstein (2022, preview) When Words Coalesce-A Sequel: "The focus of this paper lies with the development of morphemes into neo-morphemes [through fusion] [sic] [...]", cf. ch. 1.4: "The indications are, in fact, that nonfinite verb forms like participles favored the adjacency of preverb and ensuing nonfinite verb form,"

NHG Sonne géht ùnter.

NHG *Die úntergèhende Sonne.

Although this is convincing on the surface, it cannot prove the situation of PIE. gehen is not reconstructable with certainty. Allegedly

The accentuation is nonstandard – despite orthographic he there is no phonemic nor underlying aspiration, as far as I know. Alegedly, "The form gēn instead of gān is of Bavarian origin" en.WT which is rather gay (see below for Geikel vel sim.). The orthographic convention may stem from Nuremburg Kanzleisprache, perhaps. The paradigm's reconstruction is uncertain, actually (NB: Kroonen (EDPG 2013) suggests ga-prefix and cognate Lat. eo "I go"). Ad hoc, cp. past participle untergegangen, En. undergone, but see p.p. 𐌲𐌰𐌲𐌲𐌰𐌽𐍃, prefixed 𐌿𐍆𐌰𐍂𐌲𐌰𐌲𐌲𐌰𐌽𐍃 (ufargaggan, "transgress", as probably calqued in übertreten), perfective 𐌲𐌰𐌲𐌰𐌲𐌲𐌰𐌽 (gagaggan) - or rather commitative, 'to have come together'. Der Mond ist aufgegangen, might therefore imply new or full moon, passover, der Mond auf'gangen's, but the situation with prefixes is historically vexed (cf. Grimm). On the other hand, cp. Gackern (wie die Hühner), zusammen glucken (wie Glucken, Nesthennen), today chiefly sarcastic; ignoring cackle (which see en.WT, "4. A group of hyenas") as possible loan, Kluges law may apply in view of the ideophone nature of gackern, meaning gossip among womenfolk is implied ("1. Futile or excessively noisy talk", op. cit.). Admittedly, Der Mond auf'gangen's is at best obsolete, but morgen-s (at break of dawn) is fidel (ie. Sonnenaufgangs); English working nights (understood to be plural), German nachts arbeiten, Nachtschicht show the same pattern – Schicht "shift" may belong with Hauptmann "captain" to camping, speaking of the night ward (as per sanft, Nl. zaacht, En. soft).

Clitization in case of 's and 'n, which may be audible as long consonant das's /das:n/.

Assimilation helped by palatization in case of is, iʃ < ist; regressive assmiliation and compensatory lengthening in case of French est /ɛ/, liaison /ɛ.t‿/. The inflection of *h₁es- has been suggested (by whom) to show clitical pronouns, in particelar 1stP. *-mi ~ *-mos, therefore it may be questionable if the predicate attaches directly to root *h₁s- (cp. deixis Dat ass, That's fire, but This sh*t goin hard, etc.; Ger. Dieser Arsch (cp. Norse er / ert rather than -az), Das's heiß beside Es ist heiß "it is hot" (dummy pronoun) ...)

Fusion, as the title already suggested, in case of -s and -n as conservative feature of PIE, though it may be unclear how the ending derived.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.