Astori Davide Nonne interlinguistica linguisticae generalis ancilla? Some reflections on the theoretical implications of a heterodox educational path

"The heterodox branch of Linguistics", as succesfully defined by Martinet thirty years ago (cf. Martinet 1989), Interlinguistics has found its place in the University of Parma, since 2018, with an independent course (cf. official website of the last year: https://www.unipr.it/ugov/degreecourse/193578).

Articulated on the three pillars (as exemplified by the manual written specifically for it - see Astori 2020a), the course dedicates a lot of time in the classroom activity to discuss linguistic creation and creativity, starting from the founding premise of Alessandro Bausani, which concluded his famous Bausani 1974 with theese words: “At the end of this long excursus on linguistic invention we would like to conclude by affirming that any invented language, be it sacral, or of a practical type, or of a playful / infantile type or other, is an approach to the problem of language. It is a disassembling and reassembing the elements of language, trying and re-trying the various pieces in new structures, like a child playing with his Meccano. Is this distinct form the finding of the primordial One in languages (Böhme, Comenius, etc.)? Perhaps not, because the One, in language, is not, as believed in ancient times, a substance, but rather the functioning itself, a dynamis, and it is to this dynamis that all the inventors of languages pay homage". Starting from this point of view, we present, among other things, within the relationship between 'natural' and 'artificial' (see Astori b), some examples of texts, which allow us to present theoretical contents and the practical implications of interlinguistic reflection on the more general basic linguistic education (for an introductory framework see Astori c).

This contribution aims to exemplify the meaning of this educational path through some of its most emblematic didactic moments, including: Maraini’s "Le fanfole", to understand the relationship of autonomy existing between the morphosyntactic and the lexical levels; Dothraki and Na'vi as a reflection on the relationship between languages-cultures-identity-Weltanschauungen and the Sapir-Whorf hypothesis (Astori 2011); Esperanto, to understand, on the one hand, the generative plan of the language (Astori 2015) and, on the other, to introduce the themes of linguistic policies, linguistic ecology and protection of languages / cultures (see Astori 2020d,e); Markuska, for ta "poetic" approach to invented languages (see Astori 2018) by a great interlinguist to whom the University of Parma has dedicated a specific conference (see Astori 2019a, Mazza 2021); Tokipona, to think on hypeonims and on the search for semantic primitives; Marani's Europanto, to discuss the dynamics and processes of intercomprehension (Astori 2008); pasigraphies, to raise awareness of issues related to writing systems (see Astori 2021); Solresol, for the concept of denial by opposition (see Astori 2019b); Balaibalan, to underline the possible link between languages and religions (see Bausani 1954).

Bettega Simone Lawrence of Arrakis and other tales of language commodification: on the necessity of a “meta-sociolinguistics” for fictional universes

Science-fiction author Mimi Mondal has recently written that the process of worldbuilding is “inherently political”, also because “in a world where all fundamental laws can be rewritten, it is also illuminating which of them aren’t” (Mondal 2019). When it comes to conlanguages, the above could be rephrased as follows: in a world where language-building is becoming an increasingly common activity, because of the ever-growing needs of certain sectors of the entertainment industry, it is interesting to investigate why is it that some major productions appear unwilling to follow this trend (and what it is that they have been up to instead). Using as an example various types of linguistic materials drawn from four extremely popular fictional universes (Star Wars, Game of Thrones, the Marvel Cinematic Universe and Frank Herbert’s Dune), in this presentation I will address two main issues, namely (1) the “orientalizing” (Said 1978) tendencies inherent in the linguistic construction of many contemporary fictional worlds and (2) the commodification of linguistic resources (Heller 2010) often operated in these same contexts. I will argue that the sociolinguistic analysis of conlangs, and of the universes in which they exist, works simultaneously on two parallel levels, one in which the indexical values (Blommaert 2005) of any given linguistic act are to be understood within the context of the fictional universe, and one (a “meta-level”) in which the same act receives meaning outside of their imagined universe, i.e. because of the way it interacts with, and is perceived in, the real world in which we live. We, as the final consumers of such creative products, find ourselves in a vantage point from which we can discern additional meanings hidden in the speech acts of the characters that inhabit the conworlds (meanings that, crucially, are not available to the characters themselves, because they do not exist in their reality). This second level of analysis, I hope to demonstrate, is precisely the reason why a sociolinguistics of constructed universes is of paramount importance: not only because, as Jaworski and Thurlow (2010: 255) have written, the globalized societies in which we live today have created “the need for a sociolinguistics or discourse analysis that is better able to account for the hybrid, the trans-local, the spectacular, the idiosyncratic, the creative, and the multi-modal” (and, it seems to me, conlangs could be argued to fall precisely in these categories); but most of all because, by looking at the way in which linguistic resources are imagined and deployed in fictional contexts, we may learn much about the real-world linguistic attitudes of those who produce and consume said works of art.

Bobeck Dominique Root- or Stem-Based Approaches to Templatic Morphology? Evidence from Adûnaic

Even though Adûnaic is one of the less elaborate conlangs by J.R.R. TOLKIEN, it is still well structured and exceeds other languages by TOLKIEN as Entic or Khuzdul in details and corpus size by far. As he himself states, Adûnaic was deliberately constructed to resemble Semitic morphology (TOLKIEN & GILSON 2007: 85; TOLKIEN & TOLKIEN 1992: 415) with specific emphasis on a root-and-pattern system. For Semitic languages, it is still unclear whether its morphology operates on words or abstract roots. Consider the following derivations of the proposed root *nbr in Old Ethiopic (Gǝˁǝz; cf. DILLMANN & BEZOLD 2005; LESLAU 1991). Note that all “infinitives” are actually in the 3rd person singular.

(1) Verbal and nominal derivation in Gǝˁǝz

(a) nabara ‘to sit, stay, dwell, inhabit, continue’ (past tense)

yǝnabbǝr ibd. (present tense, indicative)

yǝnbar ibd. (present tense, subjunctive)

(b) ˀanbara ‘to place, put, set, establish, settle’ (past tense)

tanābara ‘to be placed’ (past tense)

(c) nabāri ‘(participle to nabara), inhabitant, steadfast, servant’

nabir ‘obstinate’

nubāre ‘position, conduct, arrangement’

manbar ‘chair, throne’

The forms in (1a–b) represent verbs of the root *nbr, on the one hand different tenses and moods, on the other hand verbal derivations such as the causative and the middle/reciprocal stem. (1c) shows several nominal derivations. But what is the morphological basis here? Traditionally, the root is combined with both a syllabic or prosodic pattern – the template – and the specific vowels. With this base, other prefixes and suffixes can combine (cf. MCCARTHY 1981, 1982; MCCARTHY & PRINCE 1990a,b; DAVIS & TSUJIMURA 2014: 192–199). More recent approaches take templates as epiphenomenon within phonology, be it root-based enhancing abstract roots (ARAD 2005, WALLACE 2013, KASTNER 2019) or word-/stem-based denying them (BAT-EL 1994, 2003; USSISHKIN 1999, 2005). For Adûnaic, these two approaches shall be proved. In contrast to Semitic languages, Adûnaic roots also contain a characteristic vowel, cf. kalab ‘to fall’ vs. kulub ‘edible vegetables that are roots not fruits’ (TOLKIEN & TOLKIEN 1992: 415f). Adûnaic derivations of the same root can have different shapes as demonstrated by the root *gimil: gimil ‘stars’ (collective), gimli ‘star’ (singulative), igmil ‘star-shaped figure’ and the unattested but possible forms gēmil, gimēl, gimmil, and gimmila. Should these roots be considered as abstract *klb+a, *klb+u, *gml+i, or are there proper words as kalab, kulub, and gimil where the derivations are based on? Or are bare stems the morphological base – being less abstract than the proposed triconsonantal roots but more than the respective words? To answer this question, parts of Adûnaic inflection and derivation shall be analysed within a Stratal OT framework, comparing the root-based approach to a stem-based one.

Chiusaroli Francesca, Monti Johanna and Sangati Federico Da Emojipedia a Emojitaliano: su un esperimento di creazione della lingua internazionale degli emoji

L'assunzione del repertorio emoji come 'lingua universale per immagini' si scontra con la naturale vaghezza semantica propria ai pittogrammi. La disponibilità di una vasta serie di segni iconici internazionali presenti nelle tastiere dei dispositivi digitali non comporta l’automatica condivisione del senso, vuoi per i divari interpretativi dovuti alle differenze culturali, vuoi per la caratteristica valenza polisemica delle immagini, per la loro incapacità di significare idee e concetti in modo univoco, per le plurime proprietà denotative e connotative delle icone, per la loro ricchezza evocativa, peculiarità che ostacolano la mutua comprensione per l’assenza di presupposti di predicibilità o di delucidazione semantica e/o morfologica.

Oltre la struttura nomenclaturista ed enciclopedica del repertorio emoji si pongono alcune operazioni di standardizzazione promosse a livello coordinato da organizzazioni impegnate in operazioni di codifica (Unicode Consortium, Emojipedia), o esperimenti che utilizzano il principio della traduzione con lemmatizzazioni e imposti abbinamenti.

In tale prospettiva e in modo originale e unico rispetto ad analoghi esperimenti, si pone l’esperienza del progetto Emojitaliano (https://www.treccani.it/vocabolario/emojitaliano_res-2f30d44e-89c2-11e8-a7cb-00271042e8d9_%28Neologismi%29/), nato nel 2016 con la traduzione collettiva del Pinocchio collodiano, proseguito con altre traduzioni, finalizzato a organizzare strutturalmente il linguaggio emoji in forma di lingua, sistema condiviso dalla comunità 'parlante'.

Ispirato a modelli e programmi storici di lingue artificiali e ausiliarie a statuto universale, Emojitaliano consiste in un repertorio di corrispondenze lessicali stabilizzate e nell’assegnazione di ruoli grammaticali nonché nella definizione di una struttura sintattica semplificata che consente di riconoscere le parti del discorso permettendo in tal modo l’orientamento all’interno del testo, l'automatica derivazione dei valori lessicali e morfologici, la decodifica del senso. Emojitaliano è, concretamente, la “grammatica” più il “glossario” di Pinocchio in Emojitaliano, ovvero il set di regole predefinite e il repertorio concordato di corrispondenze italiano-emoji, contestualmente depositate nel collegato dizionario e traduttore digitale @emojitalianobot.

Il presente intervento illustrerà i meccanismi e gli strumenti con cui Emojitaliano si costituisce come una lingua-scrittura leggibile in tutte le lingue del mondo, realizzazione del programma universalista dell’emojilingua.

Clarke Jim and Vincent Benet The 'dirty world' of translating invented languages: 'A Clockwork Orange' in Turkish

Anthony Burgess’s 1962 novella A Clockwork Orange is one of the most popular speculative works of fiction of all time, having been translated over fifty times into more than thirty different languages, as well as being adapted for stage, and filmed cinematically on at least three occasions.

Burgess's work is a first-person narrative, presented in the slang or idiolect of its protagonist, the ultraviolent teen gang member Alex. As such, Alex's argot permeates almost every line of the text. Each translator of this work is therefore faced with the challenge of adapting this invented anti-language, Nadsat, into their target language, and some translators have been more successful than others in this regard.

Since 2016, a team of researchers based at Coventry University in the UK have sought to explore how translations of an invented language may help to illuminate the underlying translation strategies of individual translators. As invented languages are not organically acculturated, and often present imaginative or speculative culturation, they pose a unique challenge to translators in this regard.

The Ponying the Slovos project has to date resulted in over a dozen published outputs, examining translations of 'A Clockwork Orange' into languages such as French, Spanish and Greek. It has also featured on the BBC and in the Guardian newspaper, and is the subject of a popular research blog, similarly titled Ponying the Slovos, which examines the topic of invented languages in popular culture more broadly, and which has welcomed nearly 30,000 readers to date.

To date, our research has focused on Indo-European target languages. However, Turkish offers an interesting expansion of our remit, as it is not an Indo-European language but yet Turkish culture is highly cognate with the cultures of those other translations we have examined thus far.

In this paper we will present the results of our examination of the Turkish translation of 'A Clockwork Orange', which has been conducted following the same categorisation process as before, and incorporated into our growing parallel translation corpus of the novella. Additionally, we will make reference to the Turkish cinematic adaptation of the novel, entitled 'Cirkin Dunya', or 'Dirty World', which emerged shortly after the well-known English language cinematic adaptation by Stanley Kubrick.

Cohen Ariel Can we construct a language without metaphors?

Metaphors are ubiquitous in natural language; in fact, no language without metaphors is known. But is such a language even conceivable? Can we construct a language without metaphors? Ever since Aristotle, scholars have debated the nature of metaphor. An issue that has proved crucial to this debate is the relation between metaphor and simile. Some (e.g., Fogelin 2011) argue that metaphors are a type of (elided) simile, or that similes are a type of metaphor (e.g., Stern 2000), whereas others (e.g., Davidson 1978) claim that the two phenomena differ in kind. Ideally, a way to settle the dispute would be to find a language that contains similes but not metaphors: this would constitute strong evidence for the latter view, namely that the two phenomena are distinct. The actual existence of such a language has not, to my knowledge, been demonstrated. But is it even conceivable? Can we construct such a language? I am not aware of any full-fledged constructed language without metaphor; but two science fiction authors have attempted to conceive of and describe just such a language. David Brin, in The Uplift War (Bantam 1987) creates birdlike aliens, who speak Galactic languages, which are claimed to have similes but not metaphors: “Unlike similes, which compared two objects, metaphors seemed to declare, against all logic, that unlike things were the same! No Galactic language allowed such nonsense.” However, Brin reproduces utterances of these aliens, in which he reflects their avian nature; and he often does so by metaphor, e.g., “The eggs of the Earthlings’ defeat had been laid so many years before.” China Miéville, in Embassytown (Del Rey, 2012), also creates aliens whose language

is claimed to have similes but not metaphors. However, these similes are frozen, and are actually idioms: they have a fixed, conventional meaning. For example, the phrase like the girl who ate what was given her is defined to mean ‘an expression intended to invoke surprise and irony, a kind of resentful fatalism’. Later on, the aliens acquire metaphors, and only then do they begin to have, in addition, novel similes: “I'm so tired I lie as still as the dead, I'm like the dead. I'm so tired I am dead”. Hence, despite their unquestionably vivid imaginations, both Brin and Miéville fail: the former constructs a language that has both similes and metaphors, whereas the other constructs a language that initially has neither, and later acquires both simultaneously. Science fiction authors, even when they construct an alien language, are still humans, and are bound by the constraints of human languages. Hence, their failure to construct a language that contains similes but not metaphors tells us that the two phenomena are fundamentally very close. To the extent that it is possible to construct a language without metaphors, it appears that it would have to do without similes too.

Colonna Valentina and Antonio Romano La lettura di poesia in lingue inventate

Lo studio prosodico della poesia, condotto in diverse realtà internazionali con metodologie autonome, ha messo in luce in questi ultimi anni la presenza di una ricca tipologia stilistica nella prosodia di questa forma letteraria (vedasi, tra gli altri, Meyer-Sickendiek et al. 2017; Puff, 2015; MacArthur et al., 2017 e Colonna, 2021). Tuttavia, restano ancora da esplorare le modalità di codifica ritmicointonativa nella lettura della poesia in lingue inventate. Questo contributo mira a prendere in esame e valorizzare il legame tra prosodia e sintassi che emerge nella resa orale di testi con tali caratteristiche. Per lingue inventate intendiamo, naturalmente, tanto le “lingue artificiali” o planned languages o constructed languages (Libert, 2018) quanto – più nello

specifico – quelle che, pur restando affini a lingue naturali, manifestano elementi di fictional language, le cui caratteristiche principali sono il non-sense e il raggiungimento di uno straniamento simile a quello della violazione delle relazioni logico-testuali (Baglioni, 2009). Per questa ricerca sono stati individuati 4 testi tratti dalle opere di 4 diversi autori, italiani e stranieri, rappresentativi della composizione in lingua inventata. In particolare sono stati scelti testi originali di Alfredo Giuliani, Fosco Maraini, Julio Cortázar e, infine, la traduzione italiana del Jabberwocky (il Mascellodonte) di Lewis Carroll. Di ciascun testo sono state raccolte due letture, una maschile e una femminile, a cura di attori professionisti. È stata impiegata la metodologia VIP (Colonna, 2021) per offrire un inquadramento generale delle letture e, considerata la peculiarità prosodico-testuale, si sono esaminati principalmente questi due nodi cruciali della relazione tra livello sintattico e lettura ad alta voce: strategie prosodiche in coincidenza di punti dubbi sul piano sintattico (con rilievo di modalità discriminanti); strategie prosodiche adottate dai diversi locutori in punti sintatticamente critici. Le descrizioni sono supportate da immagini di spettrogrammi, grafici realizzati secondo il protocollo AMPER (v. Romano et al., 2014) e, laddove necessario, da alberi sintattici supposti. Dallo studio dei materiali selezionati è risultata forte la necessità di un connubio dei differenti approcci per affrontare la materia e, seppure il corpus sia alquanto limitato, è stato possibile condurre un’analisi qualitativa che ha messo in luce diversi casi di convergenza nelle interpretazioni sintattiche, seppure con specifiche modalità intonative diverse. A ciò aggiungiamo i casi in cui le scelte sono invece state radicalmente differenti e quelli in cui invece la convergenza è totale tra le due voci, a indicare una molteplicità di soluzioni possibili e registrate. L’analisi fonetica si è dimostrata un supporto fondamentale alla comprensione sintattica e alla comprensione del testo, tale da risultare un prezioso strumento discriminante in alcuni casi. La ricerca consente un primo confronto e un primo approfondimento del tema: una valutazione più completa e un approccio quantitativo consentirebbero di fornire un quadro più ampio della questione, laddove la base di dati potesse essere più ampia, che ci proponiamo di sviluppare in futuro.

Corino Elisa and Merlo Roberto Conlang come scopo, Conlang come mezzo. Fare linguistica inventando lingue

Quando all’inizio del corso si chiede agli studenti “cosa fa un linguista?” o “quali sono le applicazioni pratiche della linguistica nella vita quotidiana?”, nella migliore delle ipotesi otteniamo risposte che riguardano la lessicografia, la definizione di norme da parte dell’Accademia, in pochi - sparuti - casi vengono citate alcune applicazioni informatiche. La vaghezza che contraddistingue l’idea che gli studenti ai primi anni di corso hanno della figura del linguista è strettamente legata alle considerazioni sulla disciplina, spesso sentita come astratta, distante, poco saliente perché avulsa da un contesto pratico e quotidiano. Si nota immediatamente la mancanza una motivazione forte, che sottenda l’acquisizione di conoscenze, ma che sia soprattutto di impulso alla creazione di competenze.

In questo contributo presenteremo la sperimentazione di un percorso didattico volto a stimolare la motivazione ed elicitare processi cognitivi alti attraverso procedimenti induttivi di analisi delle strutture linguistiche. Strumento pedagogico privilegiato di questo percorso sono le conlang.

Le conlang (constructed languages), a differenza delle lingue naturali oggetto usuale della linguistica, sono sistemi artificiali creati con vari scopi legati per lo più all’intrattenimento, spesso come parte privilegiata del conworlding (la creazione di mondi finzionali; eminente il caso di Tolkien). Dalla letteratura al cinema, alcuni dei più notevoli esempi di conlang sono le lingue elfiche Quenya e Sindarin dei romanzi di Tolkien o il Klingon creato da Marc Okrand per il franchise di Star Trek, o ancora, più di recente, il Dothraki e il Valyriano ad opera di David Peterson per Games of Thrones, che negli anni hanno guadagnato una consistente base di interessati e appassionati e sono diventate oggetto di studio anche linguistico.

Gli studenti di oggi sentono più famigliare lingua di Aragorn, di Daenerys o di Mr. Worf che quella di un parlante di tagalog o di hixkaryana.

Ecco allora che le conlang possono essere utilizzate nella didattica della linguistica in vario modo e a vari livelli, e il loro valore per i linguisti va oltre quello di semplice intrattenimento: si rivelano un buon viatico per introdurre la disciplina e possono di fatto essere la chiave per stimolare negli studenti la motivazione latente e l’interesse nell’analisi scientifica del linguaggio , così come per provocare un riflessione attiva e produttiva sui meccanismi interni ed esterni del mutamento linguistico e sulla relazione tra diasistema linguistico e diasistema socio-culturale.

Ci sono già stati, soprattutto negli Stati Uniti, alcuni esperimenti didattici incentrati sulla linguistica attraverso il conlaging (si veda ad esempio Sanders 2016); Pearson (2017), ad esempio descrive un progetto di scoperta “creativa” della tipologia e della variazione linguistica basata sull’invenzione di una lingua per insegnare agli studenti la natura degli universali implicazionali e delle strutture marcate. In Italia si tratta invece di un terreno relativamente inesplorato.

Scopo della sperimentazione è usare le conlang per creare dei percorsi induttivi di scoperta di elementi strutturali e descrittivi. Confrontando alcune battute scambiate tra cavalieri dothraki i discenti potranno isolare i morfemi lessicali da quelli flessionali, scoprendo cos’è un morfema libero e cos’è un morfema legato prima ancora che venga loro fornita la definizione. O ancora, costruendo parole e frasi del quenya scopriranno che cos’è il duale, come si usa l’allativo o come è possibile cambiare classe grammaticale usando suffissi derivativi, o confrontando quenya e sindarin osservare processi di mutamento.

Decifrare una lingua conosciuta ma mai analizzata è per gli studenti un’operazione investigativa coinvolgente e gratificante, per il docente diventa un modo per coinvolgere cognitivamente la propria classe attraverso tecniche e buone pratiche da tempo in uso nella didattica delle lingue.

Una conlang opportunamente costruita è non solo un'opera d'arte, ma anche una dimostrazione di comprensione di come funziona la lingua.

Cotugno Francesca L’impatto della tradizione linguistica classica sull’opera di Tolkien

Con il presente paper si vuole analizzare l’impatto che la lingua latina ha avuto nella produzione letteraria di J. R. R. Tolkien. La conoscenza del latino è certamente nota da parte dello studioso che è entrato in contatto con lo studio di questa lingua a partire dalla giovane età e sotto spinta della madre. Tolkien ha cominciato a studiare il latino (insieme al francese) grazie a sua madre che lo ha cresciuto all’interno della fede e cultura cattolica (Letters 66, 340, 354). Fin dalla giovane età Tolkien ha dunque associato un livello diafasico alto e connesso con la sacralità alla lingua latina, affinando poi la propria competenza in questa lingua grazie agli studi svolti presso la King Edward’s School a Birmingham. Nella lettera del 1955, inviata ad Auden (Letters 214), Tolkien stesso spiega come il Quenya si basi sia sul latino e il greco che sul finlandese. L’influsso di queste lingue nella produzione tolkeniana non è stato monolitico in realtà è mutato nel tempo, finendo con il

favorire la lingua latina, che più rispecchiava la funzione aulica e sacrale che Tolkien attribuiva evidentemente al Quenya. È dunque possibile ricostruire non soltanto una stratigrafia linguistica e un’evoluzione delle

scelte glottopoietiche dell’autore, ma identificare una serie di elementi linguistici fortemente caratterizzanti e che spaziano dalla grafematica, alla fonologia fino a toccare aspetti della morfologia.

Dedè Francesco Tolkien and comparative historical linguistics: some insights from the earliest works on Elvish languages

It is a well-known fact that J.R.R. Tolkien’s training as a philologist had a very strong impact on his work as a creator of languages. In recent years, the topic of the relationship between the languages invented by Tolkien, especially Quenya and Sindarin, and the historical languages which were Tolkien’s main source of inspiration has received new interest due to the publication of Tolkien’s linguistic material (among the most recent works, see e.g., Gilson 2020).

In this paper, I would like to address the topic of Tolkien’s ‘sources’ (on the appropriateness of such a perspective despite explicit statements by Tolkien himself, see the papers in Fisher 2011) from a different point of view, investigating the relationship between Tolkien’s earliest linguistic texts (especially Tolkien 1995; 2011) and the works in comparative historical linguistics which are likely to have been read by him and, most of all, those works which were arguably part of his training in historical linguistics. Among recent works on Tolkien’s readings, see the recent book by Cilli (2019), which will be a starting point for finding Tolkien’s relevant readings in historical linguistics.

By comparing Tolkien’s technical approach and treatment of various linguistic issues (e.g. the structure of roots in his dictionaries), the aim of the paper is twofold: on the one side, it hopes to achieve a better understanding of Tolkien’s earliest conception of his Elvish tongues. On the other side, the comparison may represent a step towards acknowledging Tolkien’s place in the history of linguistic thought.

Enguehard Guillaume Teaching linguistic methods with conlangs

As a teacher in phonology and historical linguistics, I suggest that teaching linguistics methods with conlangs can be a better choice than teaching them with natural languages. Like any scientific discipline, linguistics is based on an research object and a method. Both are usually taught to students simultaneously through material from various natural languages. However, this approach faces a major constraint: the exercises given to students are pre-set diagnoses of complex phenomena and they do not allow students to experience the difficulties of real fieldwork analysis. In a supervised context, students thus tend to repeat common patterns without really mastering the analytical methods. Without being confronted with all the practical limitations of formal analysis, students are unable to apply the method in a reallife situation. One could think that the ideal solution consists in basing a course on the detailed analysis of a single language, but students' linguistic knowledge is often limited and the presence of a graphic and/or grammatical norm often interferes with the objective analysis of the language. In this context, since 2019, I have resorted to the construction of artificial languages whose first advantage is to allow the establishment of complex, easily accessible and renewable data for which students have no extra-linguistic preconceptions. This data must have the same regularities and irregularities as a naturally formed system. To do this, it is necessary to proceed in two steps. First, students are asked to construct a list of random units (e.g. words and/or affixes). Secondly, they run this list through the filter of pre-established change rules whose role is to set up dialectal variation and distributional constraints. Irregularities due to interdialectal borrowing and data collection conditions are easily simulated by the inevitable errors of students during the changeover phase. This method seems to me to be successful in phonology and historical linguistics. In phonology, students face the chaos of data to find minimal pairs, corelations, complementary distributions and neutralisations. In historical linguistics, they browse a multitude of languages to find some equivalence rules, to establish kinship relations and proto-language roots. Students better grasp the abstract concepts of these two disciplines when they experience both the emergence process of a language and the analysis method. They are more committed to their work when they are in the uncomfortable position of not knowing if there is a solution for the task they have been assigned. And finally -- especially in historical linguistics -- they have some satisfaction in comparing the results of their analyses (the reconstructed roots) with earlier phases of the material used. In conclusion, I suggest that separing the research object and methods in teaching linguistics can be a better way of experiencing what analysing linguistic data really is. In some aspects, the practice of the method is more accurate when facing disordered artificial data than when facing ordered natural data.

Enguehard Guillaume, Lampitelli Nicola and Luo Xiaoliang Constructing languages to explore theoretical principles

Linguistic research is organized into three approaches: inductive, abductive and deductive. Induction consists in proposing general principles based on specific data. Abduction aims to relate observed data to an already existing general principle. Deduction makes it possible to check the abductive step by defining a set of expected but yet unobserved data. It can be noted that the deductive approach has a rather limited place in this heuristic process. It only serves as a method for validating the abductive step. In other words, the consequences of the principles induced by observation are only ever explored on the scale of a particular phenomenon, never on the scale of the whole system. This limitation of the deductive approach generates another issue consisting in not questioning phenomena that are so regular that a corresponding principle is implicitly accepted as self-evident. We believe that these two issues are due to the fact that the linguist has preconceived views about the object he is studying. He then remains blind to certain connections between linguistic phenomena. To deal with this issue, we propose a method inspired by experimental archaeology. In archaeology, the causes of an artifact are regularly outside our empirical field and it is not possible to make abstract generalizations as we do in linguistics. Researchers therefore sometimes use speculation and validation of hypotheses through direct experience. The experiment consists in seeing if one obtains an artifact similar to the one observed using this or that construction method. This approach is particularly famous for elucidating questions related to the lithic industry during prehistory. Though it never gives definite answers, it makes it possible to (in)validate hypotheses on the scale of a complete system. An equivalent approach applied to linguistics would be the generation of linguistic systems based on explicitly formulated principles with a minimum of typological parameters. The origin of these principles does not really matter. They may be the result of data observation, cognitive, behavioral, functional or even a priori speculation. Their value is ultimately determined only by the degree of closeness between the results obtained and the natural languages as a whole. Trying to generate similar systems pushes the linguist to explicitly define the principles that are needed and to explore all the consequences of these principles. The exploration and exploitation of linguistic principles in the construction of languages has been present since the beginnings of modern linguistics. The strong regularity of the first international languages illustrates the notion of linguistic system that emerged in the 19th century. Then, the 20th century brought an explosion of fictional languages that explore almost all facets of linguistic diversity. Even today, languages are emerging that are testing the limits of certain linguistic principles, as Toki Pona does with polysemy. Most of these initiatives address real scientific questions but from a non-academic point of view. The fact that Saussure's brother was an Esperantist is an amusing illustration of the parallel development of academic linguistics and language construction.

Gonzalez Carolina Teaching linguistics with conlanging: Language birth, Language Death

As Sanders (2016) and Punske et al. (2020) show, conlanging offers several pedagogical advantages for teaching linguistics both at the introductory and advanced levels. Conlanging is an engaging way to introduce linguistics to students who have never encountered it; and it is also an effective tool to teaching typology and cross-linguistic universals, particularly at the undergraduate level (Pearson 2020). In this paper, I discuss an approach to teach an introductory linguistics course connecting conlanging to language endangerment and revitalization that I developed and which I have taught annually at my university since 2015. The specific course is titled ‘Language birth, language death’. It is geared to incoming students, and it is part of the general education curriculum at the university. For the most part, students in this course lack a linguistics background, although many are aware (or love) conlangs such as Dothraki, Na’vi, and Sindarin and Quenya. The course fulfils several university requirements, but most of the students register because they are interested in the course topics (conlanging and/or language endangerment, or language more generally). The course begins by discussing language diversity in natural languages, and previews language endangerment and the emergence of pidgins and their development into creoles. A brief story of conlanging follows with a discussion of conlangers’ motivations and main conlang types (artlangs, engelangs, philosophical languages…). The course features a semester-long artlang project conducted in teams. This allows for the introduction of key linguistics areas (phonology, morphology, syntax, sociolinguistics and semantics) as students build their artlang from the ground up. The artlang needs to be both creative and typologically plausible. In fact, the artlang project involves building a fictional world including both a fictional map and (after discussion of phonology) the development of a writing system. The final part of the project is a translation (with glosses) of an English text connected to the fictional world. The second part of the semester focuses on the factors leading to language loss and death, the reasons why we, as global citizens, should care, and how language specialists and activists attempt to bring dying languages back to life. The discussion is enhanced by viewing a documentary on language revitalization (such as We Still Live Here by Ann Makepeace) and one or more movies that incorporate conlanging and bear on language power dynamics and/or (potential) language endangerment, such as Avatar or Arrival. Teams are asked to reflect on whether the artlang they are developing is a majority or minority language in the fictional world, and whether it is endangered. Students also reflect on the constructed aspects of revitalized languages, including Modern Hebrew and Patxohã (Macro-Jê, Brazil), and on the insights that language invention can offer to communities interested in reversing language loss (Romaine 2011). My presentation will include examples of artlangs from this course (shared with permission), some of them published in a university magazine showcasing undergraduate research and creative work, as well as examples of student feedback on the course.

Gonzales Simon Implementing Natural Language Processing and Visual Networks for Identifying Literature Landmarks on ConLang Research

Introduction and Proposal: The field of constructed languages is a growing area of study. There have been great developments from different fronts, ranging from arts and philosophy (Okrand, 1985) to more logic approaches (Nicholas & Cowan, 2003). With more acceptance as an academic field in the last decades, there have been great developments on linguistics that have contributed positively to our understanding of constructed languages (Adams, 2011; Conley & Cain, 2006; Destruel, 2016; Okrent, 2010; Peterson, 2015; Rogers, 2011; Rosenfelder, 2010). However, identifying academic research pertinent to constructed languages is a big undertaking due to the difficulty to distinguish between works that focus on conlangs and those that refer to conlangs or its keywords in a more tangential way. For this reason, this paper aims to develop a computational/visualisation framework that allows us to understand the main body of literature that is making the most relevant contribution to the field of conlang. This, at the same time, allows us to both understand the path (past and present) of this area of study, and also analyse the seminal works that have

contributed the most to our understanding and development of conlangs. These two separate tasks can be combined to create the adequate infrastructure to distinguish between studies that focus on the advancement of conlag as a field from other studies that mention conlang keywords. By separating these, we can build stronger literature frameworks that enhance the field. This, in the end, helps us identify what are the pivotal years when the field has made the greatest progress and also the literature landmarks that have made the strongest influence.

Methodology and Results: The data was obtained from the search functionality in Scopus (Elsevier) and we used a combination of constructed languages keywords (e.g. constructed, artificial, auxiliary languages) and linguistic fields (e.g. grammar, phonetics, syntax). To restrict the search to studies on linguistic approaches (excluding other areas such as engineering and psychology) the search was restricted to only Social Sciences and Arts and Humanities journals. The result was a total of more than 2,300 academic papers, from 1927 until 2022. For each paper, we analysed three main sections: title, abstract, and keywords, which we classified by the implementation of Natural Language Processing techniques to identify relationships between authors, years of publication, and work titles. From these relationships, we then created network visualisations that allow us to see connections and patterns in these three parameters. Results show that the years 2020 and 2021 have experienced an explosion of research on constructed languages, with 2019 being a year connecting more to previous decades. This may be due to great advances in computational linguistics and related fields. Authors' results show that most of the body of literature gravitates around four main authors: Noam Chomsky, Blanke Detlev, Timothy G. Reagan, and Rudolf Carnap, and it has been demonstrated that they have made invaluable contributions to the field. These results show that the methodology of this paper is reliable for the proposed analysis.

Hietaranta Pertti Constructed languages and L2 translation: comparing conlang translation with natural language translation

My main tenet here is that translation into and from conlangs (in short, conlang translation) can actually teach us something about the nature of translation in general. What follows is a partial elaboration of some the arguments this tenet rests upon. Let us start with the fact that there are, obviously, no native speakers of conlangs in the sense that there are native speakers of natural languages like English or Arabic, for example . From this it, equally obviously, follows that we cannot ask a native speaker whether a translation from, say, English into a given conlang is correct or whether there is perhaps something about the translation that needs to be amended. Given that this is so, how do we verify the correctness of a translation into a conlang so that we can accept the translation? This is of course part of the wider question of how we verify the correctness of translations in general, including natural language L2 translations, because one does not always have a native speaker available to verify the correctness of an L2 translation in the case of natural languages, either. But, as argued e.g. by Reynolds et al. (2020), the acceptability of a translation depends on a far more complicated constellation that just its correctness. Secondly, L2 translation is sometimes shunned on the basis of the fact that the cultural background which we all have as a kind of safety net for our native language skills and which supports us in our L1 translation work in a number of ways is not available to the same extent or in the same form for L2 translation as for L1 translation. Nuances tend to be harder to handle in L2 translation than in L1 translation. Given that this is so, how can we make sure that we can reach a sufficient level of subtlety and refinement in a given L2 translation into a conlang? Thirdly, nonfiction translation typically calls for terminological consistency. Is terminological consistency ever an issue to pay attention to in conlang translation? Or are conlangs perhaps safer in this regard with no terminological variation like "hard disk" and "fixed disk"? Finally, how do conlangs compare with natural languages with regard to language change? Are conlangs more susceptible to change than natural languages in view of the "rigidity and the inertia of a natural language" (Leone 2019: 11)? If they are, is this reflected in conlang translation somehow and do the different paces of change in conlangs and natural languages perhaps show in the impression the translation makes on the target language audience? These are the questions I seek to shed some light on in this presentation.

Irurtzun Aritz Ritual Languages as Natural Conlangs

Over the last decades intensive research has been done on language evolution and change, uncovering some of the biases that humans act upon when acquiring their languages. Experimental work in this area is generally done with the use of conlangs and the iterated learning paradigm (Kirby, 2017; Raviv et al., 2019). I propose a characterization ofritual languages as ‘natural conlangs’, comparable to the ‘laboratory conlangs’.

I present the Ritual Language DataBase (RLDB), which provides a typology of such languages around the world. A comparative analysis of the RLDB uncovers patterns and the significant role of users, functions, and alternate languages when shaping the grammatical structure of ritual languages. I designed the RLDB, which at the moment comprises data from 242 linguistic practices related to supernatural rituals across the world. It systematically documents 46 features of each ritual linguistic use (function; user, use of an alternate language…). The variables show different degrees of association. These are summarized in Figs. 1 (for [-Alternate language] practices) and 2 (for [+Alternate language] practices). Numbers report the uncertainty coefficient (Theil’s U), a measure of conditional entropy between variables.

Malamatidou Sofia Creative Reshaping or Linguistic Conservatism: What happens when Nadsat meets other languages?

A well-documented feature of Nadsat, the invented language in Burgess’ novella A Clockwork Orange, is that it is the result of language contact between English and Russian. However, existing research into the translation of Nadsat seems to completely ignore this feature, focusing simply instead on whether or not the foreign element of Nadsat is reserved. This reduces the translation of Nadsat, and by extension other invented languages, into a mere transfer activity of linguistic elements. What is completely ignored is the key role that translators play in linguistic innovation, as well as the motivating factors behind their creativity. This paper examines the Greek translation of the Clockwork Orange from a language contact and language change perspective and creates for the first time a link between adaptation, as understood in contact linguistics, and creativity in translation. The focus is on how Russianderived nouns in the English version have been rendered in Greek, and how these differ from naturally occurring Russian loan nouns in Greek. Results suggest that the translator’s level of creativity is a result of an interplay of factors that is related to the role of translation, the source text, as well as the target text. The paper also reveals how concepts from the field of language contact are particularly effective in reframing the way in which creativity is viewed in translation studies.

Mazzoleni Beatrice Maria Sociolinguistic features in Tolkien’s constructed languages

When Tolkien conceived his constructed languages, sociolinguistics was not yet established in the UK. Nevertheless, Tolkien anticipated some key themes of language variation and change, that would have been at the core of sociolinguistic analysis through the second half of the 20th century. Moreover, he devised language features which reflected biological and cultural characteristics of the speakers. In Quenya, the influence of the immortality of the Elves is seen on the very structure of the verb tenses, being the way Elves deal with time in speech influenced by their perception of it. Considering social aspects, Tolkien imagined a situation of diglossia, with a high form of Quenya called Parmaquesta, and a lower called Tarquesta. This distinction issued when the old Parmaquesta was substituted in the daily speech by the Tarquesta, making the fictional history of Elves enter their language conception. Moreover, both Quenya and Sindarin have dialects, which were devised from the very beginning of Tolkien’s Conlang conception. Also differences in pronunciations among different groups of Elves were devised to create a life-like scenario, as with the z/r distribution in different ethnic groups (Parma Eldalamberon, n. 17-18). Even Elvish politics had a role in shaping language: Feanor’s followers never accepted the þ > s evolution and made their speech a point of pride (The Peoples of Middle Earth). The influence of culture and spiritual beliefs can be seen also in the use or avoidance of specific expressions compared to English, which is interesting considering that, according to Tolkien, Men and Elves are two races of the same species (Letter n. 153). At the word level, an interesting case is the one of mushrooms, which are not eaten by Tolkien’s Elves. If there is a word for mushrooms in the “Quenya Lexicon”, later in the Quenya as presented in “The Etymologies” is listed only one word for fungi and sponges. In some natural languages as Finnish and Swedish there’s such a generic term and the populations didn’t eat mushrooms even in time of famine (Svanberg and Lindh, 2019). To understand further the extent of Tolkien’s social and linguistic sensitivity, also geography and legends attributed to the fictional land came to play a role in shaping the meaning of words, as in the interpretation/re-interpretation of the place name Tarlang’s Neck in Sindarin and Common Speech. Moreover, Dwarves are stated to have a special way to utter Common Speech, that Tolkien tried to render in a semi-phonetically spelled English in the Book of Mazarbul. Ents, whose impossible language and the way they reshape Quenya are only sketched in a sort of game with the reader, further demonstrate how Tolkien’s languages correspond to the nature of the speakers. As Tolkien’s Middle-earth and the tales he wrote proved to be an environment for his languages to be ‘alive’ (see Letter n. 165), so his creatures were conceived first of all as speakers of their languages, and their language was meant to fit their biological and social characteristics.

Meloni Laura, Bernhard Appelhaus, Cesco Reale Semantics of Visual Languages: the IKON language and linguistic resources

Written language production evolved from words to emoticons, to emojis, and finally to stickers (Konrad et al., 2020). The main question is: how do we realize the passage from spoken and written verbal language to a visual one? It seems a complex but possible semiotic operation. In this regard, we have successful experiments: Emojiitaliano for the translation of Pinocchio (Chiusaroli et al., 2016), Augmentative and Alternative Communication (AAC), and other modern semasiographies. These systems rely on different semantic approaches to translate concepts into pictograms or icons (Albacete et al., 1994; Tenny, 2014). The contribution of this work is threefold: we describe different semantic approaches in modern semasiographies; we then present IKON, a new iconic language conceived as a novel step into the evolution of visual languages and semantic implication of using this system in complex sentences (e.g., in the medical discourse); finally, we aim to introduce in the community engaged in the development of iconic languages the use of lexical resources and semantic ontology to build a consistent system. IKON language follows linguistic principles: consistency, completeness, semantic compositionality, and more. Up to now, IKON has a core set of around 500 icons covering basic concepts (Reale et al., 2021). This set of words is growing as meanings stem from other meanings that are semantically different concepts, whereas in natural languages they can be expressed with one polysemic word. IKON icons can be classified into a finite number of forms that allow a high degree of flexibility in visually representing complex or abstract concepts. When developing icons, abstract concepts become a more challenging task than concrete things. In these cases, we introduce the theory of Semantic Frame (Fillmore, 2003), a helpful framework when transferring a concept into an iconic system. A semantic frame carries information about the different syntactic realizations of the frame elements and their semantic characteristics. Individuating the core elements of a frame is particularly important if disambiguation of meanings is needed. When it comes to verbs, it is also essential to assess the semantic types and thematic role in the argument structure to include the appropriate participants and frames in the related visual representation. We present some case studies from the IKON language. Most lexical resources contain a large amount of linguistic information that can be exploited: Wordnet (Fellbaum, 1998) (semantic lexicon with definitions and lexical relations), FrameNet (Boas, 2005) (offer an extended amount of semantic and syntactic phrase specifications), Sketch Engine (Kilgariff et al., 2014) (a multilanguage annotated corpora resource). These resources are becoming available in other languages. Other multilingual resources are also growing (Boas, 2005; Baisa et al., 2016). Thus, a more typological approach is possible and recommended.

Meluzzi Chiara The phonological and lexical distribution of Klingon <tlh> [tɬ]

This talk deals with a peculiar sound of Klingon language, the voiceless alveolar lateral affricate [tɬ], transcribed as <tlh> and characteristing of Klingon ‘alien’ phonology. Indeed, the name itself of Klingon is tlhIngan, and many basic lexemes contain the <tlh> sound. In this work we will analyze the lexical distribution of <tlh> in the Klingon lexicon by starting with the fundamental Klingon dictionary (Okrand 1992) and the following guide to Klingon varieties (Okrand 1997), by finally taking into account the last dialogues in Klingon from the TV-Series Star Trek: Discovery. It would be demonstrated how, originally, [tɬ] could be understood as an evolution from a previous coronal [t], whereas the affricate realization has become the more standardized variant and it has also become quite productive in the creation of new Klingon words. Klingon is the artificial language created for the homonymous population of the Star Trek franchise. After their first appearance in The Original Series (TOS) in 1967, the first lines of spoken Klingon (or ‘Klingonese’ as it was called) were recorded for 1979 film Star Trek: The motion picture. However, it was only for the third 1984 movie Star Trek: The Search for Spock, that Klingon characters have to speak many lines in their own language. It was, thus, decided to develop a proper language, with its own grammar and vocabulary, and the linguist Marc Okrand was called up to the task (Okrand 2011). Nowadays, Klingon has claimed to be the largest fictional language, and his speakers could be count around hundreds, with a linguistic and cultural association (the Klingon Language Institute, KLI) and language certificates (see Okrent 2010). Its use is also quite attested in popular culture and fans’ autonomous productions (Meluzzi 2019). In order to sound ‘alien’ but still be pronounceable by human actors (Okrand et al. 2011: 115), Klingon presents a peculiar phonology. Despite having a small phonological repertoire, with 21 consonants and only 5 vowels, it’s characteristic of Klingon the presence of many velar, uvular and glottal sounds, often presented together within the same word. And, obviously, affricate [tɬ] strongly characterizes this language, and it is phonologically quite rare, since it could be found in a few Northern and Central American languages (e.g., Cherokee and Nahuatl, from which it has probably been taken, Okrand 1992: 15), and in other minority languages across different families, like Tswana, a Bantu language, Chukchi, and Gherdëina Ladin). Despite its peculiarity, in the first 1985 edition of the Klingon dictionary, <tlh> occurs only in 62 words over 1,500 entries. It usually occurs in monosyllabic word with the basic CVC structure, and more likely as C1. Only in two cases <tlh> is found both as C1 and C2 of the same word, i.e. the onomatopeic words tlhutlh ‘to drink’ and tlhetlh ‘to progress’. More words with <tlh> occur in followup of Klingon language, thus demonstrating the key role of this sound within Klingon phonology and

contributing in its ‘alien’ sounding (cf. Devenny 2017). Other interpretations are possible and will be further discussed.

Salazar Rodó Sergio José and González Carolina The Impact of Formal Instruction on the Practice of Conlanging

Individuals typically begin constructing languages as a self-taught hobby (Peterson 2015). This is because conlanging requires a steep learning curve due to the complex-linguistics knowledge required. The advent of the internet and the abundance of online conlanging resources such as Zompist (Rosenfelder 2021) have been critical to lowering this curve while facilitating the development of conlanging communities such as the Language Creation Society. In conjunction with the increased visibility of constructed languages due to popular media, the surging popularity of conlanging in the 2010s prompted the creation of academic courses teaching how to construct languages(Punske et al. 2020). This opened a new field of formal instruction in a subject that had historically been self-taught. Given the high barrier-to-entry in linguistics knowledge required to create a language, this has resulted in the creation of some courses targeted at students with limited-to-no linguistics background (Anderson et al. 2020, Berry 2020).

My experiences with conlanging began as a self-taught hobby until I participated in such a course on language invention. Although this liberal-arts course was among those targeted at individuals with no experience, I was able to derive much knowledge and experience from the exposure to structured education on the subject, co-operative work, and formal presentation. Building upon my contemporary reflections of this course, this project explores my experiences and correlates to them existing literature on the instruction of conlanging itself in order to consider the value and effectiveness of formal conlanging education in those who already carry experience (Sanders 2016, Peterson 2020). This will be accomplished through directly comparing established formal intervention techniques and goals with those presented in the course and relating them to my own growth and the reported growth of students in the literature.

In particular, the evolution of the specific process of conlanging that I use because of the course is emblematic of the type of impact instructional interventions can have. Therefore, special consideration will be given to discussing the different structures of the process of conlanging available to the general public and comparing them to those presented in formal courses (Rosenfelder 2010, Peterson 2015). This will be deconstructed down to individual elements, including goal setting, starting point (phonology, morphology, lexicon, etc.), practice exercises (if present), and approaches to naturalism. Direct comparison between these elements as present in literature and in my own contemporary approaches to language construction thus allows for an assessment of how much the various sources, both informal and formal, have impacted my conlanging. All of this allows for the opening of a critical discussion over the overall benefits to formal conlanging intervention, even in individuals with stated experience.

Thalmann Christian and Peterson David J. Aldjen Helt: Creating the Fjerdan Language for Netflix' Shadow and Bone

In recent years, more and more cinema and TV productions have commissioned conlangs to enrich their worldbuilding and deepen the audience's immersive experience. One such production is «Shadow and Bone», the Netflix adaptation of Leigh Bardugo's Grishaverse book series. Here we showcase our work on the Fjerdan language for this show, offering a behind-the-scenes view of how we analyzed the corpus of attested language samples from the books, built the grammar and vocabulary, and translated lines of dialogue for the show.