Expression of Several Grammatical Meanings in Oral vs. Graphical Constructed Languages

Roman Viktorovich Tarasov

Department of Applied and Experimental Linguistics, Leo Tolstoy Higher School of Russian and Foreign Philology, Institute of Philology and Intercultural Communication, Kazan Federal University, Kazan, Russia

Abstract

This article attempts to discuss three grammatical meanings: tropative (‘X considers Y to be Z’), apparetive (‘X seems to be Y’) and causative in different types of constructed languages, or conlangs. While most of conlangs, as well as natural languages (NLs), possess an oral form (and thus are called oral), the others only have a written form based on semantics (and thus are graphical). Conlang typology has a low level of coverage. However, results of such studies might be extremely useful for general linguistics. The reason is that the derivational models of conlangs are chosen by their creators according to the aims of the languages. Thus, the language authors express their own judgment on difficulty/simplicity of language structures. So, the regularities of conlang typology might be used for explaining the typological universals. This project is the first attempt at studying conlangs with the same method as NLs, namely, the cross-sectional method described in my article (Tarasov 2021). The models of twenty oral conlangs (OCLs) are compared with those of four pasigraphies. Although the ranges of two samples are incommensurable, these are all the pasigraphies that are available for examination. The results of this article show that while in terms of tropative and apparetive, OCLs and pasigraphies behave differently (with both being different from NLs), they have much more in common both with each other and with NLs in terms of causative.

Keywords: Grammatical Meanings, Constructed Languages, Pasigraphies, Typology

(To Download PDF File of This Full Article in full Details, Click Here)

1. Introduction

My article is dedicated to the means used for expression of three grammatical meanings (tropative, apparetive and causative) in constructed languages, or conlangs. Conlangs are languages that appear as a result of a deliberate creation, whereas natural languages (or NLs) are a result of a natural development. Conlangs can be oral or graphical. While oral conlangs (or OCLs) possess a spoken or signed form, graphical conlangs (or pasigraphies) have only a written form and express meanings directly without any phonology. Conlangs are usually divided into four categories according to their aim:

– Auxiliary languages (auxlangs), which are designed as lingua franca for native speakers of different languages. All pasigraphies belong to this category.

– Zonal auxlangs (zonlangs), which are designed as lingua franca for a particular language family or area.

– Artistic languages (artlangs), which are designed for the works of art or as pieces of art themselves.

– Engineered languages (englangs), which are meant to check linguistic hypotheses

The following grammatical meanings are to be discussed:

• tropative — ‘X considers Y to be Z’. X is a subject, Y is an object, Z is a characteristic.
• apparetive — ‘X seems to be Y’. X is a stimulus, Y is a characteristic.
• causative — ‘to cause smb/smth to be or to do smth’

It is necessary to emphasize that while most of conlang-related papers concern cultural or psychological aspects of language construction, this article studies conlangs with purely linguistic methods, i.e., the same way as natural languages (NLs) are usually treated.

Despite the fact that conlangs are almost always ignored in typological research, there are some strong arguments in favour of conlang typology studies. Firstly, typology is quite a young branch, while conlangs have been created since the 12th century. This means that the choice of a model could only be based on the creator’s own position on what is easy or naturalistic and what is not in a language, i.e., on metalinguistic reflection. Data about conlangs can help us explain linguistic universalities and diachronic changes. Moreover, authors of pasigraphies have to express their own judgement on which sign is appropriate for expression of a particular meaning, i.e., their metasemiotic reflection. One more argument is that sometimes it is difficult to draw a border between natural languages and conlangs, e.g.:

• Basic English by C. Ogden, Newspeak by J. Orwell can be regarded as controlled versions of an NL (namely, English);
• Similar phenomena exist within Slavic language family called microlanguages (e.g., West Polesian by M.Shelyagovich, Siberian by Y.Zolotarev, etc.) — a class introduced by A.Dulichenko (1981) and involving varieties of Slavonic NLs having an author but based on several natural varieties of one or two NLs.
• There is a problem of reconstructed languages. Both Modern Indo-European and Hebrew belong to this category, but the former language is considered to be a conlang, while the latter is always regarded as natural.

2. Literature Review

Conlang grammar and typology is a field which is still in its infancy. Conlang sometimes becomes a focus of a linguistic paper, but are mostly approached to from the points of conlanging philosophy (Piperski 2017; Sidorova & Shuvalova 2006) or L2 acquisition (Windsor & Stewart 2017; Hracs 2011; Carpenter 2005, 2006, 2010; etc.). There are still some papers about conlangs as a thing-in-itself, but they are mostly dedicated to auxlangs, e.g. (Libert 2010, 2014).

There were some attempts at including (oral) conlangs into the natural typological research projects as a separate subsample, e.g. (Tarasov 2019, 2021a) on typology of tropative. A conlang subsample consisting of 13 oral conlangs is also represented in the HSE Tropative Database (HSE 2020) providing information on tropative systems of 183 NLs. The second example is (Tarasov 2020) on typology of negative concord and single negation. A conlang subsample consisting of six oral conlangs is also represented in the HSE Negative Concord Database (HSE 2021) providing information on participant negation patterns in 39 NLs. The papers (Tarasov 2021b, 2022) are dedicated to lexical derivations and participant negation, respectively, in OCLs. The results of these projects are published in the HSE Conlang Derivations Database (HSE 2022) and the HSE Conlang Negation Database (HSE 2023).

Pasigraphies receive even less linguistic attention than OCLs. Namely, it would hardly be an overstatement to say that they do not receive it at all. More often such languages are examined within Speech Pathology Studies (Dada et al. 2012; Harty et al. 2005, etc.) or Design (Nawar 2016). L2 Acquisition is also a typical field for such papers: (Alant et al. 2009; Raghavendra, Fristoe 1995; Jennische, Zetterlund 2015).

Pasigraphies are not represented in typological research, but Tarasov (2022) made an attempt to include them into the OCL typological project on participant negation patterns. Four pasigraphies constitute a detached subsample in (HSE 2023).

3. Methods of the research

For this project, two research methods were used.

For studying tropatives and apparetives, the cross-sectional method described by Tarasov (2019, 2021a, 2021b) and involving translation of six sentences requested from users of constructed languages, was applied. Sentences for translation were offered in Russian or English. The list of sentences was as follows:

1. I consider him to be intelligent
2. He is considered to be intelligent
3. I do not consider him to be intelligent
4. He is not considered to be intelligent
5. He seems to be intelligent
6. He does not seem to be intelligent

In (Tarasov 2021), the disadvantages of this method are admitted and described. Firstly, it does not give us an ability to exclude an existence of a particular model in a language. This risk can be omitted by marking constructions as ‘detected’ or ‘not detected’ rather than ‘present’ or ‘absent’. Moreover, this method does not allow us to omit a risk of an informant’s mistake. However, grammar description analysis cannot be applied for this part of research since tropative and apparetive are rarely mentioned in language descriptions. Using a larger questionnaire or offering more complicated sentences for translation will decrease a number of participants willing to cooperate. The main advantage of a method is that it allows to examine more languages than others. Moreover, it was an aim of the research to compare derivational models of conlangs with those of NLs. Thus, methodological uniformity should have been provided.

However, for studying causatives, grammar description analysis was used. Causative is a meaning that is usually sufficiently described by language creators, thus, it can be examined with such a method. The cross-sectional method is inappropriate for this part of research, since causative studies require examining a much wider range of contexts. Thus, a large questionnaire would be required, which contradicts the principle of the cross-sectional method.

I also have to admit that I had to use grammar descriptions for studying all the meanings in two of four pasigraphies (the Nobel Universal Pictorial Language and Paleneo) since they have no active community but possess detailed descriptions.

Overall, 19 OCLs and four pasigraphies were examined.

4. Results: tropative and apparetive

4.1. Terminology

Tarasov (2021a) introduces the notions of positive and negative tropative. While a positive construction have a meaning of personal opinion (‘X has an opinion that Y is Z’, ‘X considers Y to be Z’), a negative construction denotes absence of opinion or of an opposite opinion: ‘X does not consider Y to be Z’. By analogy, I introduce the notion of positive and negative apparetive: ‘X seems to be Y’ (‘X makes an impression of being Y’)  vs ‘X does not seem to be Y’. If a negative tropative or apparetive in a language is a grammatical negation of a positive one, such a model is positive-negative symmetric. Positive-negative symmetric apparetive model can be called just symmetric, because there are only two types of apparetive constructions.

In the same article, the notions of direct and reverse tropative are introduced. While a direct construction introduces the subject X having a particular opinion (‘X has an opinion that Y is Z’, ‘X considers Y to be Z’), a reverse construction only expresses the existence of an opinion about an object: ‘Y is considered to be Z’. Thus, direct tropative is a triadic predicate T (X, Y, Z), whereas reverse tropative is dyadic T'(Y, Z). If a reverse tropative is a grammatical passivization or detransitivization of a direct one, such a model is called direct-reverse symmetric. If a tropative model is both direct-reverse and positive-negative symmetric, it can be called absolutely symmetric. There can be no reverse apparetive, because this predicate A (X, Y) is intransitive — it only has two arguments: a stimulus X making an impression Y.

All four types of tropative constructions and both types of apparetive constructions constitute a questionnaire offered for informants.

Four degrees of tropativity are also introduced in the same article. By analogy, Tarasov (2021b) introduces three degrees of apparetivity.

1^st degree tropative or apparetive is expressed morphologically (with an affix). It is strong if applicable to any stem of a particular class and weak otherwise; it can also be either polysemic or monosemic. Tarasov (201b) also offered to regard tropative or apparetive expressed ‘with a copula’ as 1st degree ones, but there is no sufficient proof in this paper that the Lojban model having this feature (and this model in general) is different from usual analytical models. Thus, such models will be regarded as analytical in this paper.

For example, Arabic has a 1st degree tropative model: hasuna ‘to be good’ — ist-’ahsana ‘to consider good’ [Jacques 2013: 1]

The conlang called Klingon a 1st degree apparetive model: val ‘to be intelligent’ — vallaw’ ‘to seem intelligent’ [Tarasov 2021b: 6]

2^nd degree tropative or apparetive is expressed syntactically (with one finite clause), but not morphologically. It can also be polysemic or monosemic.

For example, English has 2nd degree models: I find him smart; He seems smart. The same is true for the majority of other languages, including Persian: Man u-rā hušmand hesāb mi-konam

3^rd degree tropative or apparetive is expressed lexically but not syntactically. All the necessary arguments can be stated explicitly. Thus, 3rd degree tropative construction is not a triadic predicate T (X, Y, Z), but rather a combination of two diadic predicates:

1. T (X, O): X has an opinion O
2. O (Y, Z): Y is Z (X’s opinion)

3^rd degree apparetive construction is not a dyadic predicate A (X, Y) but rather a combination of a dyadic and a monadic one:

1. A(I): There is an impression I
2. I (Y, Z): Y is Z (an impression)

Such examples exist almost in all languages, e.g., I think he is smart; It seems that he is smart.

4^th degree tropative is descriptive only. Tropative subject is not stated explicitly, thus, there is no distinction between direct and reverse tropative.

Such examples are probably universal. Jacques (2023) rejects Tarasov’s (2019, 2021) statement that Arrernte lacks tropative providing an example of a tropative construction:

• the Margie            lhwarrpe         are-me

1sg.erg          Margie            sad.abs           see-n.pst

‘Margie looks sad to me’ [Jacques 2023: 3]

Language is prescribed the highest degree of all models detected in it. If a model is detected futher, a degree can be upgraded (or assigned if a language was wrongly regarded as tropativeless or apparetiveless).

(To Download PDF File of This Full Article in full Details, Click Here)

4.2. Data discussion: tropative

The tropative studies brought the following results. None of OCLs registered in the sample utilizes 1st degree models. However, the majority of them (11 out of 19) possess 2nd degree. Six languages belong to the 3rd degree, and the other two only use 4th degree models.

Out of 17 languages belonging to the 3rd or higher degree, eight tropative systems are direct-reverse symmetric, and nine are asymmetric (in seven cases direct constructions are used instead of reverse ones, in the other two independent constructions are used). 18 of 19 languages are positive-negative symmetric, while Solresol is the only language with double negation marking (depending on the attitude towards the characteristics Z, either a verb milado ‘to praise’ is used for positive constructions and substituted with an antonymous verb dolami ‘to scold’ in negative constructions or vice versa).

Typical auxlang model is 2nd degree (see examples 1.2 and 1.3 from Esperanto; with a single exception of Lidepla, utilizing a 3rd degree model). Also, four out of six systems are absolutely symmetric, while Solresol is the only positive-negative asymmetric conlang in a sample, and Lidepla is direct-reverse asymmetric (see examples 4 and 5).

• mi (ne)     opini-as           li-n                  sağa    homo

1sg      (neg)  consider-pres 3sg-acc          smart   person

‘I (don’t) find him/her smart’ [Tarasov 2019: 8]

• li (ne)     opini-at-as                  sağa    homo

3sg      (neg)  consider-pass-pres    smart   person

‘(S)he is considered to be smart’ [Tarasov 2019: 8]

• me opini                ke        ta         es         intele

1sg      consider.pres comp   3sg      cop      intelligent

‘I think that he is intelligent’ [Tarasov 2021b: 4]

• oni opini                ke        ta         es         intele

3pl      consider.pres comp   3sg      cop      intelligent

‘He is considered to be (lit. They think that he is) intelligent’ [Tarasov 2021b: 4]

Zonlangs have much in common with auxlangs: three out of four zonlangs utilize 2^nd degree models, while pan-Nigerian Guosa use 3rd degree models. International Sign is the only direct-reverse asymmetric zonlang in the sample (in this language, direct and reverse constructions are independent):

• 1sg see 3sg intelligent

‘I consider him to be intelligent’ [Tarasov 2021b: 5]

• 3sg to.have.reputation intelligent

‘He is considered to be intelligent’ [Tarasov 2021b: 5]

Artlangs use totally different structures. 2nd degree strategies were not detected in any of four artlangs. Moreover, in one of them, namely Dothraki, only 4th degree model was detected:

• Me nem     nesa     fin                    yotnhare          mae    

3sg      postp  known conj.anim      brain                3sg.poss

haj-a

strong-3sg

‘It is known that his brain is strong’ [Tarasov 2021b: 6]

The case of engelangs is similar to those of auxlangs and zonlangs: three languages out of five utilize 2nd degree models. Toki Pona, a minimalistic engelang, has the 3rd degree of tropativity, while in a gynocentric conlang called Laadan only 4th degree model was detected. Ithkuil tropative is unique in the way that X is an indirect object of a direct tropative construction marked with a relative case:

• Thuzaleoč üode

intelligent.3sg 1sg.rel

‘He is intelligent, according to my opinion’ [Tarasov 2021b: 7]

Tropative verbs can be polysemic, the range of primary meanings is quite wide:

• milado ‘to praise’, dolami ‘to scold’ (Solresol, auxlang, the earliest in the sample)
• ay ‘to say’ (Sambahsa, auxlang, modern)
• kol- ‘to consider’ (Globasa, auxlang, modern)
• imeti ‘to have’ & uvažati ‘to respect’ (Interslavic, zonlang, pan-Slavonic)
• find- ‘to find’ (Folkspraak, zonlang, pan-Germanic)
• g-L-U-v inessive marker + proximity marker + mind (aUI, englang, oligosynthetic)

The same study carried out for pasigraphies brought absolutely different results. Out of four languages, two (Paleneo and the Nobel Universal Pictorial Language) use morphological-level tropative models (morphological level is analogical to the 1st degree). Moreover, Paleneo uses implicit tropative, i.e, marks tropative with zero:

• ‘good’, ‘to accept’
• ‘bad’, ‘to reject’ [Chateris 1972: 31]

 It is unknown whether this model is strong or weak since Paleneo has no active community and Charteris (1972) also does not provide such information.

The NUPL tropative system is also unique in the way that in this language, tropative is compatible with dynamic verbs, like ‘to come’:

2sg      can.neg           trop-come      horse   run                   fast

                        cond  2sg pres.neg    neg.keep 3sg.inan    graze

‘You can’t expect the horse to run fast if you don’t let it graze.’ [Randic 2009: 394-395]

It is also unknown whether the NUPL morphological tropative marker is strong or weak since this language also lacks active community and Randic (2009) also leaves this aspect unattended.

It is worth mentioning that neither implicit tropative nor dynamic verbal tropative is registered in any oral language (whether natural or constructed).

The other two pasigraphies possess lexical-level tropative systems (lexical level is analogical to the 2nd and the 3rd degrees, since syntactic criterion cannot be checked for such languages). Moreover, Blissymbols possess morphological-level reverse tropative:

• good-think.int ‘correct’
• bad-think.int ‘incorrect’ [BIC 2020]

The Mediaglyphs tropative system is analogical to that of Esperanto: lexical-level and absolutely symmetric:

• ( )  { }

1sg      (neg)  consider          3sg      intelligent

‘I (do not) find him smart’ [elic.]

•      ( )  { }

3sg      pass    (neg)  consider          intelligent

‘He is (not) considered to be intelligent’ [elic.]

4.3. Data discussion: apparetive

The results from apparetive studies were as follows. Two languages possess 1st degree apparetive models, while four other languages only possess 3rd degree ones. 13 languages belong to the 2nd degree of apparetivity. All of the systems in a sample are symmetric.

All auxlang systems are 2nd degree:

• (no) jin-om             sagatik

(neg)   seem-3sg        intelligent

‘He seems/does not seem intelligent’ [elic.]

The same strategy is utilized in three zonlangs out of four. The single exception is a pan-Nigerian conlang called Guosa. In this language, only 3rd degree model was detected:

• O/ko bika     kpe      o          di         yeze

            3sg/3sg.neg   seem    comp   3sg      cop      intelligent

‘It seems that he is intelligent’ [elic.]

Artlangs demonstrate a significant difference from auxlangs and zonlangs in terms of apparetivity, as well as tropativity. In two languages out of four, only 3rd degree models were detected. On the other hand, Klingon possesses a 1st degree apparetive model:

• val-law’

intelligent-app

                        ‘He seems to be intelligent’ [Tarasov 2021b: 6]

As for engelangs, results from this category are different from the others. Three languages out of five have the 2nd degree of tropativity, while Ithkuil and Toki Pona have the 1st and the 3rd degree, respectively. The Ithkuil morphological apparetive model looks as follows:

• tv-älo-rd-a ma

                        intelligent-state-app-3sg        3sg

                        ‘He seems to be intelligent’ [Tarasov 2021b: 7-8]

Apparetive verbs can also be polysemic, the range of primary meanings is smaller than that of tropative verbs, but still quite wide:

• aspekti ‘to be seen as’ (Esperanto, auxlang, the most famous)
• vygledati ‘to look’ & kazati si ‘to show oneself as’ (Interslavic, zonlang, pan-Slavonic)
• nef ‘probable’ (Sindarin, artlang by J.R.R. Tolkien)
• O-i-v feeling + shine (aUI, englang, oligosynthetic)

The same study carried out for pasigraphies brought slightly different results. None of four languages possesses a morphological-level model. However, this might be example with a small range of a sample (four languages, whereas the frequency of morphological-level tropative within OCLs is much less than 25%). Three languages use monosemic apparetive models, while Paleneo is the only one using a polysemic apparetive verb  ‘to be similar’. It is worth mentioning that such a polysemy is nonexistent (or at least not detected) in OCLs and rare (less than 1% frequency [CLICS 2019]) among NLs.

4.4. Conclusion

– There are no 1st degree tropative systems among 19 OCLs, but three morphological-level tropatives among just 4 pasigraphies. It is necessary to mention that such a feature is not that frequent among NLs: just 7% [HSE 2020].

– On the contrary, there are no morphological apparetives in pasigraphies. However, it might be because the sample is scarce. There is no reliable statistics on natural languages.

– Two other features not registered in oral languages are implicit tropative (found in Paleneo) and dynamic verbal tropative (found in the NUPL).

5. Results: causative

5.1. Terminology

Tarasov (2021b) describes the most common strategies used for causative expression in oral auxlangs. Firstly, causative can be expressed either morphologically (English: simple — to simpl-ify, Persian: res-id-an — res-ān-dan) or analytically (to make difficult). A morphological causative marker is strong if applicable to any stem of a particular class and weak otherwise. This marker can also be verbal, non-verbal or universal. The most common analytical strategies are:

– Non-integrating causative verbs. Causative verbs are integrating if they denote both causation and a caused action (to feed, to show, to kill), whereas non-integrating verbs only express a meaning of causation (to order, to command, to allow).

– Stem alteration: using an independent stem for a causative verb (to eat — to feed, to see — to show, to die — to kill).

– Implicit causation: using the same stem for a caused action and causation.

5.2. Data discussion

The results of causative studies were different from those of other meanings analysis. 16 languages out of 18 utilize morphological causative markers, and those markers are strong in 13 languages, while three languages only possess weak ones. In two languages, causative is only expressed morphologically. The most common analytical strategy is non-integrating verbs, which are present in 16 languages. Stem alteration is also a widespread strategy, which is present in 10 languages. Only two languages within a sample use implicit causation.

The case of auxlangs is quite interesting. Five languages out of six possess morphological causative, while Solresol, the earliest auxlang in the sample, only expresses this meaning analytically. Moreover, four languages out of the other five use strong markers with a single exception being Volapük, which is the second earliest auxlang. Thus, we can come to a conclusion that an evolution of causative in auxlangs took place with strong universal morphological causative becoming an optimal strategy for an oral auxlang.

Solresol is also one of two OCLs using implicit causation:

• simisol ‘simple’, ‘to simplify’ [Tarasov 2021b: 8]

At the same time, typical auxlang causative model looks like this (examples 22 and 23 are from Globasa):

• side ‘to sit’ — side-gi ‘to seat’
• bala ‘strong’ — bala-gi ‘to strengthen’ [Globasa 2019 4.2]

Zonlangs utilize causative systems different from those of global auxlangs. Namely, none of three zonlangs in a sample utilizes universal morphological markers. The reason is probably that such conlangs are based on NLs of a particular family, and universal markers are not widespread in natural languages. While pan-Germanic Folkspraak and pan-Romance LFN use strong markers, pan-Slavonic Interslavic utilizes weak ones. All these markers are non-verbal.

Artlangs, which are usually aimed to be unnaturalistic, use strategies that are paradoxically similar to those of auxlangs. Indeed, all four languages of this type use strong universal markers.

Engelangs demonstrate quite an interesting distribution. While three language utilize strong morphological markers, the other two only express causative meaning analytically, but there are no languages with weak markers. Such polarization might seem logical since weak markers require memorizing the range of stems compatible with them. But what is more strange, is the fact that languages with opposite aims use the same strategies. For example, Lojban was designed by the Logical Language Group to strictly follow the rules of math logic, whereas a gynocentric englang called Laadan was designed to be oriented on emotions. However, they both use strong universal causative markers. On the other hand, the englangs lacking morphological causatives are Toki Pona (designed as ideally minimalistic) and Ithkuil (aimed to be as sophisticated as possible). Ithkuil also uses implicit causation:

• atř ‘to be observable, to make observable’

The same approach applied to pasigraphies brought results similar to those of auxlangs. All four languages use morphological causative markers. Moreover, markers are strong in three languages. The NUPL is the only pasigraphy to use several weak universal markers. It is also the only pasigraphy within a sample using implicit causation:

• ‘to turn over, to get turned over’

Mediaglyphs have quite an interesting feature: in this language, a non-verbal causative marker is a superimposition of a verbal causative marker and a non-verbal inchoative marker:

• ‘real’ — { } ‘to make come true’ (caus-real)

Paleneo is interesting in the way that it has a limitless compatibility of a causative markers with the stems of different classes:

• ‘to boil’ (caus-100^°)

5.3. Conclusion

• Both oral and graphical conlangs mainly express causative morphologically with strong markers. This is also typical for NLs, only 8% of natural languages lack causative markers [WALS 2005].
• Among analytical strategies, implicit causative is the least frequent strategy — there is also no evidence that any natural language regularly uses such strategy.
• However, there is a significant difference between NLs and conlangs: in the latter category, universal markers are more widespread than strictly verbal or non-verbal.

6. General conclusion

– Causative is morphological among most OCLs and in all pasigraphies under discussion, and this is the common feature of NLs (92% language utilize morphological markers) and conlangs.

– On the other hand, three of four pasigraphies utilize morphological tropative, while none of OCLs does it and this feature is rare in NLs (7%).

– In terms of apparetive, pasigraphies also demonstrate some peculiarities, while there is too little reliable data about NLs.

– Thus, while three categories of languages have similar causative features, their tropative and apparetive features are not the same. The most probable reason is that causative is more sufficiently studied and more often expressed grammatically than both tropative and apparetive.

Bibliography

Alant, Erna; Bornman, Juan; du Preez, Anlie. 2009. ‘Translucency and Learnability of Blissymbols in Setswana-speaking Children: An Exploration. Augmentative and Alternative Communication’. 25(4), pp. 287-298. Baltimore.

BCI 2020. Blissymbols Communication International. Symbol File 8483-27158.

Carpenter, Angela. 2005. ‘Acquisition of a Natural vs an Unnatural Stress System’. In Papers from the 29th Boston University Conference on Language Development.

Carpenter, Angela. 2006. Acquisition of a Natural versus an Unnatural Stress System. Doctoral dissertation, University of Massachusetts, Amherst.

Carpenter, Angela. 2010. ‘A Natural Bias in Learning Stress’. Phonology 27: 345-392.

Charteris, Leslie. 1972. Paleneo: a Universal Sign Language. Hodder & Stoughton Limited. London.

CLICS 2019. The Max Planck Institute for Evolutionary Anthropology. The CLICS database. https://clics.clld.org/

Globasa 2019. Globasa. Grammar. https://xwexi.globasa.net/eng/grammar

Harty, Michal; Alant, Erna; Life, Henriette. 2005. ‘Comparison of the learnability and retention between Blissymbols and CyberGlyphs’. International Journal of Language & Communication Disorders. 40(2):151-169.

Hracs, Lindsay. 2011. The Acquisition of L2 Information Structure in Hungarian: The Salience of Position vs Prosody. Master’s thesis, University of Calgary.

HSE 2020. HSE Tropative Database. https://artemorekhov1999.pythonanywhere.com/

HSE 2021. HSE Negative Concord Database. https://grishakuznetsov-workshop-app-app-b3e3op.streamlit.app/

HSE 2022. HSE Conlang Derivations Database. https://conlang-database.romtarrus2000.repl.co

HSE 2023. HSE Conlang Negation Database. https://conlang-negation-database.romtarrus2000.repl.co

Jacques, Gillaume. 2013. ‘Applicative and Tropative Derivations’. Linguistics of Tibeto-Burman area. Volume 36.2. P.1-13.

Jacques, Gillaume.2023. ‘Estimative constructions in cross-linguistic perspective’. Linguistic Typology 2023: 27(1). P. 157-194.

Libert, Alan Reed. 2010. ‘Comparing Comparatives in Artificial Languages’. Interlingvistikaj Kajeroj, 1.1:64-174.

Libert, Alan Reed. 2014. A Survey of Interjections in International Auxiliary languages. Journal of Universal Language 15-1, p. 79-103.

Nawar, Haytham Adel. 2016. A Theoretical Model for the Design of a Transcultural Visual Communication System in a Posthuman Condition. PhD Dissertation. University of Plymouth.

Piperski, Alexander Chedovich. 2017. Конструирование языков: от эсперанто до дотракийского (Language Construction: from Esperanto to Dothraki). Moscow, Alpina Non-fiction.

Raghavendra, Parimala & Fristoe, Macalyne. 1995. ‘No shoes; they walked away?”: Effects of enhancements on learning and using Blissymbols by normal 3-year-old children’. Journal of Speech and Hearing Research, 38, pp. 174-188.

Sidorova, Maria Yurevna & Shuvalova, Oksana Nikolaevna. 2006. Интернет-лингвистика: вымышленные языки [Internet-linguistics: invented languages]. 1989.ru, Moscow.

Randic, Milan. 2009  Nobel universal language. 2009.

Tarasov, Roman. 2019. ‘Towards a Typology of Tropative’. Conference on Typology and Grammar for Young Scholars. Saint-Petersburg, 2019, November 21-23.

Tarasov, Roman. 2021a. ‘Grammaticalization and Lexical Expression of Tropative from a Typological Perspective’. Ahwaz Journal of Linguistics Studies. Vol. 2, No. 2.

Tarasov, Roman. 2021b. ’Tropative, Causative and Apparetive in Different Types of Constructed Languages: a Typological Approach’. Conference on Typology and Grammar for Young Scholars. Saint-Petersburg, 2021, November 25-27.

Tarasov, Roman. 2022. ‘Towards a Typology of Single Negation and Negative Concord in Constructed Languages’. Conference on Typology and Grammar for Young Scholars. Saint-Petersburg, 2022, November 24-26.

Windsor, Joseph W. & Stewart, Robyn. 2017. ‘Can Unnatural Stress Patterns Be Learned: New Evidence from Klingon’. In Proceedings of the 2017 Annual Conference of the Canadian Linguistic Association.

 

(To Download PDF File of This Full Article in full Details, Click Here)