Language of Criminal Confessions

Confession evidence is powerfully persuasive, and yet many wrongful convictions involving false confessions have surfaced in recent years (Innocence Project, 2021; National Registry of Exonerations, 2021). Although police are trained to corroborate admissions of guilt, research shows that most false confessions contain accurate details and other content cues suggesting credibility as well as extrinsic evidence of guilt. Hence, a method is needed to help distinguish true and false confessions. In this study, we utilized a corpus-based approach to outline the linguistic features of two sets of confessions: those that are presumed true (n = 98) and those that have been proven false (n = 37). After analyzing the two corpora in LIWC (Linguistic Inquiry and Word Count) to identify significant categories, we created a logistic regression model that distinguished the two corpora based on three identified predictors: personal pronouns, impersonal pronouns, and conjunctions. In a first sample comprised of 25 statements per set, the model correctly categorized 37 out of 50 confessions (74%); in a second out-of-model sample, the predictors accurately classified 20 of 24 confessions (83.3%). A high frequency of impersonal pronouns was associated with confessions proven false, while a high frequency of conjunctions and personal pronouns were associated with confessions presumed to be true. Several patterns were observed in the corpora. In the latter set of confessions, for example, “I” was often followed by a lexical verb, a pattern less frequent in false confessions. Although these data are preliminary and not to be used for practical diagnostic purposes, the findings suggest that additional research is warranted.

When one is singular: Notes on zero-person constructions in Latvian

Axel Holvoet has demonstrated that Latvian has two types of zero-person constructions which formally differ in grammatical number, and that the singular type has a parallel in Finnic, but not in Lithuanian. This paper shows that the meanings covered by the two types are distinct and do not overlap. Using the framework proposed by Gast and van der Auwera for the description of human impersonal pronouns, it is shown that the singular type is characterized by non-veridicality and an internal perspective. As in Finnish, but not Estonian, it is used in conditional sentences with all kinds of verbs. The plural type is used in Latvian as well as in Lithuanian with veridical propositions and an external perspective.

NMN-VD: A Neural Module Network for Visual Dialog

Visual dialog demonstrates several important aspects of multimodal artificial intelligence; however, it is hindered by visual grounding and visual coreference resolution problems. To overcome these problems, we propose the novel neural module network for visual dialog (NMN-VD). NMN-VD is an efficient question-customized modular network model that combines only the modules required for deciding answers after analyzing input questions. In particular, the model includes a Refer module that effectively finds the visual area indicated by a pronoun using a reference pool to solve a visual coreference resolution problem, which is an important challenge in visual dialog. In addition, the proposed NMN-VD model includes a method for distinguishing and handling impersonal pronouns that do not require visual coreference resolution from general pronouns. Furthermore, a new Compare module that effectively handles comparison questions found in visual dialogs is included in the model, as well as a Find module that applies a triple-attention mechanism to solve visual grounding problems between the question and the image. The results of various experiments conducted using a set of large-scale benchmark data verify the efficacy and high performance of our proposed NMN-VD model.

Nominal speech act structure: Evidence from the structural deficiency of impersonal pronouns

In this paper, we propose that there is a speech-act structure in the nominal spine, just as there is in the clausal spine. Its function is to encode what we do when we utter a nominal: that is, we name, describe, or track individuals. Thus, speech-act structure establishes a link between the discourse referent and the speech-act situation. The evidence we discuss comes from nominals that lack this speech-act structure, namely impersonal pronouns. We argue that impersonal pronouns have in common that they lack nominal speech-act structure but are not otherwise a natural class: they vary in syntactic structure. Thus, we propose a novel formal typology of impersonal pronouns.

Human impersonal pronouns in West Germanic

In this article, we examine and compare the main human impersonal pronouns in Afrikaans, Dutch and English. The second person singular, the third person plural and the ‘man’- and ‘one’-pronouns are studied by means of an acceptability judgment questionnaire and a completion questionnaire. The combination of the two methods reveals interesting descriptive facts about the three West Germanic languages. They include, among other things, the ‘man’-prominence of Afrikaans versus the ‘you’-prominence of Dutch and English for expressing the universal meaning ‘anyone’ and the more prominent position of ‘they’ in Dutch than in the other languages for conveying the existential meaning ‘someone, some people’. Our findings have a number of more theoretical implications too. The two existing semantic maps for human impersonal pronouns make different distinctions in the existential domain, based on type/level of (un)knownness on the one hand and number on the other. Our study tests both sets of distinctions and shows that the two dimensions interact with each other in Afrikaans, Dutch and English. The results thus support a recent proposal in the literature for a combined semantic map. The data from the completion questionnaire, finally, also indicates that existential uses prefer alternative forms of impersonalization to human impersonal pronouns in all three languages.

Tax Preparer Certification and Organization Form among Uncertified Preparers Influence Client Satisfaction and Experience

ABSTRACT How does state certification (e.g., CPA, enrolled agent [EA], or attorney) and organizational form among uncertified preparers (big box “franchise” firms versus not) influence client satisfaction and interactions? Results from a sample (n = 3,984) of Yelp ratings of U.S. tax preparers indicates that certified preparers receive higher ratings than do uncertified small-firm preparers, who earn higher ratings than do franchise preparers. Linguistic analysis suggests that clients of certified preparers experience tax preparation service as a “relational” interaction, with reviews evidencing more positive affect and inclusive pronouns (e.g., we). In contrast, the clients of franchise firms experience tax preparation as “transactional,” with reviews evidencing more swear words and impersonal pronouns. Linguistic results for uncertified small preparers fall between those of certified and franchise preparers. The paper is among the first to adapt emerging corpus linguistic methods to examine client perceptions of professional accounting services.