Evidence for a Contemporary
Language Shift
in Progress:
The English Passive
Professor
Vloor Vorth
Chair in Sociosyntactic
Syntactic Sociolinguistics
New evidence has emerged of an in-progress shift in the grammatical realisation of the English passive. Drs. Sophie Oh and Lund Guistick have been working on the ‘Let it out’ project (OMG Funded), an investigation of sociolinguistic variation in the use of ‘let’ among working class left-wing, great-grandchildren of blacksmiths, bakers or other tradespeople in rural Yorkshire. In the course of gathering data on the structure ‘let someone know’, they came across the following two utterances from J.
(1) Aye, so it’s miself what wer let known first ’bout ev’fing what were ’appening down t’road
(2) E’s bin let known, like, a fousan’ times ’bout it, but ’e ain’t never done nuffin ’bout it.
Aside from the balanced, rhythmic patterns and epic tone of these utterances, the central point as regards language change is the apparent shift in realisation of the past participle from the first verb in the multi-verb string ‘let’ to the second verb ‘know’, realised as ‘known’.
Sophie Oh and Lund Guistick were far too busy making lists of different exponents and mapping them to socio-economic background, etc., etc., etc., to bother with any analysis of this structure. However, they forward it by linguamail to Professor Zee Smith and Dr Txakx Jones—famously of the ‘What is it then that’s what might be ’appenin’ ’ere?’: Investigation into Syntactic Variability and their Impact on Democratic Robustness—who couldn’t wait to get their scholarly hands on, and obsessive minds wrapped around, it.
After six months of investigation, including a well deserved five-month mid-way rest period in Honolulu, Zee and Txakx have come up with two very different analyses of and predications around this tantalising structure. Zee’s take hinges on the morphological invariability of ‘let’. Along with a few other verbs of English, all monomorphemic and ending in [t] or [d] (put, set, cast, wed, spread, etc.), ‘let’ does not vary in its past or past participle form. Let us call these ‘monosyllabic, [t]/[d]-lexeme final morphologically invariant verbs of contemporary international English’ for short, or MT/DLFMIVoCIE for very short.
Zee hypothesises that this invariability pushes the morphological realisation of the passive onto ‘know’ in this lect, an operation she terms ‘Push the Passive Past’ hypothesis (PPP). This gives the following prediction: in any V[1] + V[2] chain where V[1] is a MT/DLFMIVoCIE verb, V[2] will undergo PPP. To exemplify, on the assumption that phrasal verbs follow the same pattern as these V[1] + V[2] chains, ‘put it down’ in this lect will become ‘it was put downed’. Alternatively, a translation into this lect of relevant portions of Scripture would read ‘the demon was cast outed’. The derivation is therefore as follows:
- Active V[1] + V[2] chain e.g. ‘let him know’
- Apply passive
- Abstract schema = let + know + passive morphology
- V[1] = MT/DLFMIVoCIE verb
- Passive realisation blocked
- Invoke PPP
- know → known
- He was let known
Txakx, by contrast, offers an analysis which draws on his seminal Feature Flow Syntax model (FFS). In FFS, abstract features can ‘flow’, under strictly delimited circumstances, between adjacent (and occasionally non-adjacent) lexemes. In the structure under consideration, the absence of the direct object which is present in the active version (i.e. ‘let him know’ → ‘he was let [ZERO MATERIAL] known’) makes the string a potential FFS candidate. In this analysis, therefore, ‘let’ does contain a passive marker zero morph but by FFS, it is duplicated on ‘know+n’. This would mean that the above example has a very different underlying structure which Txakx formalises as follows:
- Active V[1] + V[2] chain e.g. ‘let him know’
- Apply passive
- Abstract schema = let + know + passive morphology
- V[1] is MT/DLFMIVoCIE verb
- V[1] takes passive morphology as a zero morph
- Apply FFS
- Passive morphology duplicates on to ‘know’
- He was let[+ zero morph passive morphology] known
In her response to Txakx’s analysis, Zin argued that this risks overgeneralisation as non-MT/DLFMIVoCIE verbs will also realise the passive morpheme and by FFS see it duplicated onto the V[2] as MT/DLFMIVoCIE does not act as a blocking principle under Txakx’ analysis. This would predict that ‘make him work’ would surface as ‘he was made worked’. Txakx addresses this by speculating that FFS has some additional constraint or limitation on it that prevents such forms occurring.
With these two models available, Oh and Guistick have re-applied themselves to the copious data from Yorkshire in an attempt to locate other exponents which may lend support to either of the above. We look forward to further investigation into this possible exciting shift in the English passive.
