Researchers recently embarked on a fascinating journey to understand how constructed languages (conlangs) in popular fantasy culture—such as J.R.R. Tolkien’s Black Speech—evoke specific impressions in listeners. The results, published in Language and Speech, uncovered several factors that influence perceptions of conlangs, irrespective of the context in which it is presented.
“Personally, I have been a huge fan of Tolkien for a long time. And I always found it amazing how he chose the names for people or locations that were quite fitting. The languages he invented seemed to be designed in order to invoke beauty or evilness,” said study author Christine Mooshammer, a professor of German Language and Linguistics at Humboldt University in Berlin.
“So what makes words sound pleasant or unpleasant? Is it in the sounds themselves or the image we have about the speakers? There are some studies on the individual perception of natural languages with the limitation that participants can associate foreign languages with the respective cultures, in other words: the problem of stereotypes. This problem can be solved by testing languages that (most) people cannot recognize, such as constructed languages.”
“As a professor in phonetics I thought this might also interest some students and proposed a class ‘Why does Orkish sound evil?’ My expectations were really exceeded: some of the students were also conlangers who had already invented a couple of languages.”
For their study, the researchers selected a diverse array of twelve conlangs from popular fantasy literature and media — including the Lord of the Rings, Star Trek, and Game of Thrones — ensuring a range of phonetic and phonological characteristics were represented. These languages were chosen based on their fictional cultural backgrounds and the intentions of their creators regarding the sounds and overall aesthetic of the language.
The study included 129 German speakers. These participants were asked to listen to audio recordings of sentences in these various conlangs and rate them on three scales: pleasantness, goodness (versus evil), and peacefulness (versus aggression). The sentences were recorded by two speakers (one female and one male) in a neutral tone, without emotional inflection or special sound effects. Importantly, the sentences were presented without any context or information about the fictional cultures they represented, focusing the participants’ attention purely on the auditory experience.
In terms of pleasantness, goodness, and peacefulness, the researchers found that Klingon from Star Trek and Dothraki from Game of Thrones were the most negatively rated, while Tolkien’s Elvish languages Quenya and Sindarin were the most positively rated.
The researchers also found that languages with a higher percentage of voiced intervals were rated more positively across all scales, indicating that voicing contributes to a more pleasant or agreeable auditory experience. Voiced intervals refer to segments where the vocal cords vibrate to produce sound. This vibratory action is a critical aspect of producing vowels and certain consonants. Voicing is the difference between sounds like /b/, /d/, /g/ (voiced consonants) and /p/, /t/, /k/ (voiceless consonants).
An unexpected outcome of the study was the relatively positive reception of Orkish, a language intended by Tolkien to sound menacing and harsh. The findings suggested that the Neo-Orkish samples used in the study were perceived as more pleasant and less aggressive than anticipated.
“One of our main findings was that Orkish doesn’t sound evil without the special effects, seeing the speakers and hearing the growls and hissing sounds in the movies,” Mooshammer told PsyPost. “Therefore, the average person should be aware of the effect of stereotypes that do influence the perception of a language. Do languages such as German sound orderly and unpleasant and Italian beautiful and erotic because of its sound or just based on one’s own attitude toward its speakers? ”
Further dissecting the phonological and phonetic dimensions, Mooshammer and her colleagues found that the percentage of sounds not present in the listener’s native language (German in this case) had a notable impact on the ratings. Languages with a higher proportion of non-native sounds were rated more negatively, suggesting that unfamiliar phonetic elements contribute to a perception of otherness or strangeness, affecting the overall aesthetic and emotional response.
Mooshammer noted that previous research has hinted at a complex relationship between social stereotyping and sound structure. A study published in Psychology of Learning and Motivation in 2020 found a correlation between familiarity with a language and its aesthetic appeal to a listener. At the same time, however, languages with higher sonority scores were perceived as more pleasant, while those with syllable-timed rhythms were preferred over stress-timed languages, suggesting intrinsic emotional associations of specific phonetic properties.
But the new study, like all research, includes limitations. For example, the reliance on short excerpts to represent each language might not have allowed participants to fully grasp the phonetic and phonological richness of the conlangs, potentially affecting their ability to accurately assess the intended emotional qualities.
“Our sound examples were very short,” Mooshammer explained. “In our next study, we will present longer stimuli and less languages. And, in the current study, we cannot really exclude that our participants thought they recognized a language and then judged it according the their prejudices.”
Looking forward, the researchers aim to discern whether there are universal traits that determine these aesthetic judgments or if such perceptions are heavily influenced by an individual’s native language and the spectrum of languages they are familiar with.
“Our long-term goal is to find out what makes a language (natural or constructed) sound beautiful or evil,” Mooshammer said. “Are there any universal traits or does it depend on your native language or the languages you already know? And can we really separate phonaesthetics from our image of the speakers that speak this language?”
The research team is actively seeking participation in an ongoing experiment designed to explore these questions in greater depth. By collecting a broader dataset across a diverse range of languages and listener backgrounds, the team hopes to shed light on the complex dynamics of phonaesthetic perception. Interested individuals are encouraged to participate in the study by visiting https://warpgrid.xyz/conlang-experiment/.
The study, “Does Orkish Sound Evil? Perception of Fantasy Languages and Their Phonetic and Phonological Characteristics,” was authored by Christine Mooshammer, Dominique Bobeck, Henrik Hornecker, Kierán Meinhardt, Olga Olina, Marie Christin Walch, and Qiang Xia.
