Can a question be a lie? An empirical investigation

In several recent papers and a monograph, Andreas Stokke argues that questions can be misleading, but that they cannot be lies. The aim of this paper is to show that ordinary speakers disagree. We show that ordinary speakers judge certain kinds of insincere questions to be lies, namely questions carrying a believed-false presupposition the speaker intends to convey. These judgements are robust and remain so when the participants are given the possibility of classifying the utterances as misleading or as deceiving. The judgements contrast with judgements participants give about cases of misleading or deceptive behaviour, and they pattern with judgements participants make about declarative lies. Finally, the possibility of lying with non-declaratives is not confined to questions: ordinary speakers also judge utterances of imperative, exclamative and optative sentences carrying believed-false presuppositions to be lies.

5. Note that both kinds of non-declarative insincerity discussed here involve presuppositionsbelievedtobefalse.Theydifferwithrespecttothecontentofthosepresuppositions:only the second kind of non-declarative insincerity involves presuppositions about something other thanthespeaker'sattitudes.Anotherdifferencebetweenthetwokindsofnon-declarativeinsincerity concerns the kinds of presuppositions that are in play. Non-declarative disinformative insincerityinvolvesbelieved-falsepresuppositionstriggeredbyparticularexpressionsorclauses, whichareusuallyclassifiedassemantic presuppositions.Bycontrast,therelevantpresuppositions innon-declarativeattitudinalinsincerityarestandardlyclassifiedaspragmatic presuppositionsor speaker presuppositions. Throughout this paper, we will only be concerned with cases involving semanticpresuppositions.
Prediction 3: While ordinary speakers judge disinformative questions to be non-lies, they judge the declarative counterparts of such questions to be lies.
We conducted three experiments to test these predictions. Experiment 1 tests whetherordinaryspeakersconsiderdisinformativequestionsasliesorratheras misleadinganddeceiving.Theresultsofthisexperimentsuggestthatordinary speakers do in fact consider such questions as instances of lying, contrary to Prediction1.
Wethenconductedathirdexperimenttocompareverdictsaboutdisinformativequestionswithverdictsabouttheirdirectdeclarativecounterparts.This allowsustocompareverdictsaboutdeclarativeandnon-declarativeinsincerity, whileleavingallotherfactorsofthetestedcasesconstant.TheresultsofExper-iment3speakagainstPrediction3:ordinaryspeakersclassifybothdisinformative questions as well as their declarative counterparts as lies, with the more prototypical declarative lies being judged as lies only slightly stronger than insincerequestions.

Design, Procedure and Materials
Theeighttestvignettesallincludedanagentwhointendedtodeceivethe addresseebyposingaquestiontoher.Thisquestionincludedapresupposition that the agent believed to be false and that was also actually false. Moreover, theaddresseealwaysendedupwiththefalsebelieftheagentintendedtocause intheaddressee.Ascontrolcases,wealsoincludedavignettefeaturingaclear lie (phrased as a declarative sentence) and a vignette in which the agent both utteredatruedeclarativesentenceandwasfullysincere.Herearethewordings ofthreevignettes(onetestcaseandthetwocontrolcases).Allothervignettes canbefoundintheAppendix:

Exam (test case, disinformative question):
Anna and Billy are classmates. Billy has missed a few days of school because he was ill.Anna wants Billy to think they have a maths exam tomorrow,althoughthereinfactisnosuchexam.Shetextshim:"Have yourevisedforthemathsexamtomorrow?"Billycomestobelievethat therewillbeamathsexamtomorrow.

Discussion
Doestheconceptoflyingasitisheldbyordinaryspeakersincludedisinformative questions? The aim of Experiment 1 was to answer that question and, in doingso,testthepredictionofStokke'sdefinitionoflying,accordingtowhich utterancesfeaturingdisinformativequestionscannotbelies.Theresultsdemonstrate that ordinary speakers do count such questions as lies.Across all eight test cases, participants expressed that the speaker lied to the addressee when utteringadisinformativequestion.Theseresultsoccurforboththe7-pointratingscaleaswellasforthebinaryscale.Importantly,presentingpeoplewiththe possibilitytoclassifytheutteranceasacaseofmisleadingordeceivingdidnot significantlyreducelieratings.

Experiment 2: Disinformative Questions Compared to False Implicatures and Deceptive Behaviour
Thefindingsofthefirstexperimentsuggestthatordinaryspeakersconsiderdisinformativequestionstobelies.Apotentialexplanationforthisfindingholds thatordinaryspeakershaveanundifferentiatedconceptoflyingandthatthey hencedonotdistinguishbetweenlyingontheonehand,andmisleadingutterancesordeceptivebehaviourontheother.Ifthisturnsouttobecorrect,thefolk conceptoflyingwouldbemuchbroaderthanthephilosophicalconceptoflying (asoutlinedinSection2).Analternativeexplanationholdsthatthefolkconcept isnotmuchbroaderthanthephilosophicalconcept.Laypeopledodistinguish betweenlyingandotherformsofdeception,buttheynonethelessconsiderdisinformativequestionstobelies. Experiment 2 aims to decide between these two explanations by adding utterances and behaviours that philosophers consider to be uncontroversial cases of deceiving (without lying) or misleading (without lying). If the firstexplanationiscorrect,weshouldexpectthatpeopledonotdifferentiate between cases involving disinformative questions and cases usually considered as clear examples of deceiving and misleading. Both should be clearly judgedaslies.If,however,peopleonlyclearlyjudgedisinformativequestions to be lies and differentiate them from examples considered as clear cases of deceivingandmisleading,thefirstexplanationisunlikely,andwehavefound additionalevidencefortheclaimthattheordinaryconceptoflyingdoesallow for the possibility of lying with disinformative questions -without being overlybroad.
Deceptivebehavioursontheotherhandweremostlynotconsideredtobe lies, and even less so when they could also be judged as instances of deceiving and misleading (which they were). For false implicatures, responses were roughlysplit(andalsotendedtobelowerwhenadditionaloptionswereavailable), indicating that participants seemed to be unsure or undecided about whethertheyshouldbecountedaslies.However,theywerealsoclearlyjudged asinstancesofdeceivingandmisleading.

Participants
Participants were again recruited using Prolific Academics, completed an online survey implemented in Unipark and were required to be at least 18yearsold,nativeEnglishspeakersandtonothaveparticipatedinourpreviousexperiments.242participantsstartedthesurveyandthedataof180were includedintheanalysis(62wereexcludedfornotfinishingtheexperiment, failing an attention check in form of a simple transitivity task or failing a manipulation check that assessed whether the utterances in the vignettes were phrased as declaratives or questions). Mean age was 34.6 years, 31% weremaleand69%female.Participantsreceived£0.50forestimated5minutesoftheirtime(£6/h). A 7-point Likert-item, ranging from "Definitely did not" to "Definitely did",was usedto be able to detect-in contrast to the binary format-differences in the degree to which participants considered an utteranceaslying.

Design, Procedure and Materials
On the final page, participants were asked about demographics and presented with an attention check and a manipulation check ("In the majority of cases that I saw in the experiment, one person deceived another by using a [Question/Declarativesentence]").

Exam (Declarative):
Anna and Billy are classmates. Billy has missed a few days of school because he was ill.Anna wants Billy to think they have a maths exam tomorrow,althoughthereinfactisnosuchexam.Shetextshim:"There willbeamathsexamtomorrow."Billycomestobelievethattherewillbe amathsexamtomorrow.

Design, Procedure and Materials
Participantswererandomlyassignedtooneoffourconditionsina4(sentence type: Questions vs. Imperatives vs. Exclamatives vs. Optatives; between) * 3 (scenario:Exam, Mercedes, Claire's Party;within)mixeddesign.Theprocedurewas thesameasinthepreviousexperiments.Thethreevignetteswereonlyslightly adjusted to match the respective sentence type. Each participant was asked whether the speaker lied to, misled or deceived the addressee. The complete materialcanbefoundintheappendix.
A second definition of lying that eschews the notion of what is said and allowsfornon-declarativeliesisofferedbyViebahn(2017;2020;2021).Onthis view,speakershavetocommitthemselvestosomethingtheybelievetobefalse in order to lie. This commitment-based approach allows for non-declarative lies, while counting at least some cases involving believed-false implicatures (suchasstandardcasesofmisleading)asnon-lies.Thedefinitionisthusbroader than says-based definitions, while being narrower than Meibauer's inclusive definition. It seems to fit with the results in Reins and Wiegmann (2021), in which lie-judgements for a range of different utterance types pattern with commitment-judgements.