It is often repeated on the internet that studying one year of Esperanto followed by three years of German will get you more proficient in German than if you had studied it all four years. The sources to support this claim are sparse, though, and many of the studies that have been done had small sample sizes, were based on superficial or subjective metrics, lacked comparable control groups, or had poorly-defined scientific objectives. There is some information to be gleaned by carefully reading the literature that exists, though, and so I provide this summary of the state of the art in the hope that it can help readers arrive at a more grounded and nuanced understanding of this phenomenon.
In 1995, the Italian minister of public education Rosa Russo Jervolino commissioned a study on the benefits of teaching Esperanto in public schools. The commission found that Esperanto would have great benefits, both for communication and for education, citing multiple experiments as evidence, and recommended that it be provided as an elective in all schools (at least two laws were proposed in the Italian Chamber to this effect, but neither of them passed). The commission’s report was published in the official bulletin of the Ministry of Public Education (MUIR) and has become the basis for most discussion on the internet about Esperanto as a teaching language.
However, many of the experiments cited in the report did not actually test whether Esperanto helped students learn other languages – just whether it was easier to learn than other languages. In addition, most of the cited experiments were carried out by teachers, not scientists, and were published in newspapers or magazines, not peer-reviewed scientific journals. There have been several studies by linguists published in scientific journals that were not cited, and while their results do not unanimously contradict the commission’s conclusions, I find that they do paint a more nuanced picture.
Before we get into that, I will provide some definitions. When studying a particular language accelerates the subsequent study of other languages, it is said to be propedeutic, which means “pre-teaching” (the spelling “propaedeutic” is also common, but I think “propedeutic” is simpler and easier to read – and more consistent since almost no one writes “encyclopaedia” or “paediatrician” nowadays).
I will divide propedeutic value into four hypothetical types, similar to the three posited by Renato Corsetti (psycholinguistics professor at La Sapienza di Roma and president of the Universal Esperanto Association):
- General. Someone who has studied a foreign language is more aware of linguistic structures, which makes them better at learning any subsequent languages.
- Emotional. A language that is difficult to learn will frustrate and confuse students, which may cause them to give up on language learning entirely. A language that is easy to learn will make students feel competent, which will motivate them to learn more languages.
- Activity-derived. The faster a student becomes proficient in a language, the sooner they can begin to use that language in activities outside of class (such as writing to penpals, or taking other classes in that language), which allows them to spend more total time studying foreign languages, and improves their proficiency in all languages via number 1.
- Specific. A regular language can serve as a bridge between two natural languages, by introducing a student to roots and grammatical features in a sterilized environment, which helps them to learn those roots and grammatical features more quickly when they are later exposed to a messy and irregular natural language that also has them.
Note that logically speaking, specific propedeutic value must be situation-dependent. A specifically propedeutic language that helps Hungarian students learn Italian will not have the same effect if the students are French, or if the language being learned is Finnish.[2,14]
It is widely accepted that all languages possess some general and specific propedeutic value, so to set a fair bar in this regard, I will make a distinction between “strong propedeutic value” and “weak propedeutic value”. A language that has weak propedeutic value will help you learn other languages, but not enough to make up for the time you could have spent studying them directly. If Esperanto only has weak propedeutic value, then learning it may provide some inherent value, but is not a worthwhile endeavor if you just want to be good at speaking German. On the other hand, a language that has strong propedeutic value will help you learn other languages so much faster that you will save time in the long run, despite having delayed studying those other languages. If Esperanto has strong propedeutic value, then anyone who wants to learn German should spend a year studying Esperanto first, even if they have no intention of ever using Esperanto, because it will allow them to eventually surpass their peers who started with German.
With those terms defined, we can turn to the question of interest: “Does Esperanto have strong propedeutic value, and if so, by which of the four mechanisms listed above?” Various experiments have been performed over the last one hundred years, which I will summarize here.
Bishop Auckland, UK; 1921
The earliest experiment on the propedeutic value of Esperanto whose results I can find was conducted between 1918 and 1921 at a girls’ school in Bishop Auckland by one Alexandra Fischer. According to the MUIR bulletin, it concluded that Esperanto helped English students and French and German by bridging their grammars, transparently representing linguistic structures, and accustoming them to the idea of cognates. The same experiment was also described, in slightly more detail, in a 1922 report to the general secretariat of the League of Nations titled “Esperanto as an International Auxiliary Language”. That report was much less kind, saying in no uncertain terms that “the results were inconclusive. The girls who had learnt Esperanto made more rapid progress in French than those who had learnt no foreign language, but the reverse happened in the case of German.”
I don’t think we should read too much into the result of this. I find it very unlikely that the knowledge of Esperanto would make it more difficult to learn German, since it sounds like all of the test subjects studied German for the same length of time. So in the absence of more details, I suspect all they saw was the random variation that you can expect from a small sample size.
Auckland, New Zealand; 1924
In 1924, an experiment at the Bishop elementary school of Auckland, New Zealand was described in the Enciklopedio de Esperanto. I haven’t found a copy I can access, but according to the MUIR bulletin, it investigated whether Esperanto could help students learn French faster. It doesn’t say what its conclusions were.
New York City, USA; 1934
In 1934, Helen Eaton, a lexicologist specializing in Western European languages, published a peer-reviewed article in the Modern Language Journal. It details three experiments that were performed by psychologist Edward Thorndike, Columbia University professor Laura Kennon, and Eaton herself, in partnership with the International Auxiliary Language Association. The goal of the studies was to determine whether studying a general language course based on Esperanto helped students learn French vocabulary more quickly.
The conclusions are mixed. One of the experiments wasn’t actually complete when the paper was written, and Eaton is reluctant to draw any conclusions from it. Both of the others did find evidence of weak propedeutic value, but only one saw evidence of strong propedeutic value, despite there not being any obvious difference between the two experiments to explain such a discrepancy. In her report to the IALA, Eaton admits that the sample sizes were small and the control groups weren’t completely comparable with the experimental groups, and says that these studies should not form the basis for anything other than further research.
Sheffield, UK; 1952
In 1952, J. H. Halloran, a lecturer in pedagogy at the University of Sheffield, published a peer-reviewed article in the British Journal of Educational Psychology. He describes an experiment in which 40 students were taught one year of Esperanto followed by three years of French, while another 46 were taught four years of French. The groups were controlled to be of equal intelligence (measured by the Shipley Abstraction Test) and learn French from the same set of teachers. They were then tested for their skill in French by eleven different metrics.
Halloran found that Esperanto had strong propedeutic value for the students with lower Shipley-intelligence, but not for the students with higher Shipley-intelligence. When all of the students were taken together, their average French scores after the four years were the same regardless of whether one of those years was spent on Esperanto. In other words, spending a year on Esperanto helped the students who would normally have struggled in French, but for students who would normally do well, delaying their study of French was only a hindrance.
Halloran goes a step further and breaks down the score differences by metric, and finds that the Esperanto students tended to rank better than the non-Esperanto students on the metrics of passive knowledge (comprehension, vocabulary, and grammar) but worse on the metrics of active proficiency (speaking, reading, and writing).
Even though his sample size was relatively small, I admire the measures Halloran took to ensure the control group and the experimental group were as similar as possible, and I find these results more compelling than most of the others listed here.
Somero, Finland; 1963
In 1963, Joel Vilkki, a teacher at Somero High School, published an article in the magazine El Popola Ĉinio. In it, he details an experiment in which two parallel curriculum tracks were developed: one in which students studied Esperanto for one year before switching to German, and the other in which they started on German immediately. Both groups also studied Swedish, but it doesn’t say during which years. In addition, after finishing their first year, the students on the Esperanto track had their geography classes in Esperanto rather than in their native Finnish, attended at least one Junior Esperanto Congress as a class, and kept in touch with Esperantist pen-pals from around Europe.
According to Vilho Setälä, a famous photographer and patron of the Esperanto Foundation, by the third year of the experiment the Esperanto class appeared much more confident in their German lessons than the non-Esperanto class did. Vilkki corroborates this in El Popola Ĉinio, and in addition reports that at that time, the students in the Esperanto track had nearly caught up to the others in German – the Esperanto class was covering material that the non-Esperanto class had covered only one day before. Vilkki’s article doesn’t explicitly say so, but Erik von Hertzen, in his 1997 biography of Vilkki, states that the Esperanto students had managed to surpass the non-Esperanto students by the end.
Despite the study’s unstated sample size and lack of objective metrics to measure students’ proficiency, its results do seem to be compelling. At first glance, this would seem to contradict the three previous studies, none of which found conclusive evidence of strong propedeutic value in Esperanto for all students. However, there are two significant differences between this experiment and the others we have seen. First is the fact that the students’ native language, Finnish, is not related to either Esperanto or to the target language, German. This creates the optimal conditions to observe specific propedeutic value. Second is the fact that the Esperanto students used various activities outside of class to practice their Esperanto after the first year – the geography classes, the Junior Esperanto Congress, and the pen-pals. This creates a situation in which we might observe activity-derived propedeutic value. These differences make this study uniquely informative on those two types of propedeutic value, but also makes it difficult to guess which of them might have caused the effect he observed.
Manchester, UK; 1965
In 1965, Norman Williams, the headmaster at the Egerton Park County Secondary School in Denton, Manchester published a peer-reviewed article in The Canadian Modern Language Review, a journal about the teaching and learning of second languages. In it, he briefly describes seven experiments performed at Egerton Park over the course of twenty-five years.
In the first and second experiments, Williams simply tested the ease with which students could learn Esperanto, and found that it was much easier to learn than French. In the third experiment, it was found that smart children could learn Esperanto faster than average children.
The fourth experiment found that students who had studied a foreign language were better at their native English than those who just took English classes. The fifth found that students who studied Esperanto saw more of this improvement than students who studied other languages. It is not stated how students’ improvement in their native language was measured. I might be cynical, but given that it was England in the 1960s, I suspect that it was mostly their ability to memorize Latin-derived words and conform their writing to outdated English grammar rules.
In the sixth experiment, students were taught one year of Esperanto, and then the ones who performed well proceeded to four years of French. Williams concluded that Esperanto does have strong propedeutic value. However, he admits that the study was questionable on the grounds that the control group and the experimental group were not concurrent; they were separated by ten years. In that time, the teachers and teaching methods had changed significantly, which would have had a large impact on students’ performance that would be difficult to separate from the impact of Esperanto. And while Williams does not point it out, the fact that the Esperanto class was used to weed out the students who performed poorly surely inflated the average final scores of the students who studied French. Williams assures readers that “no matter how one looked at the results, there was ample justification for believing that one year of Esperanto followed by one year of French, on the basis of progress in French, was better than two years’ French without any introductory course in Esperanto.” However, given that he does not elaborate this claim, I must disagree. At best, the data are suggestive; I don’t see how an experiment without a comparable control can possibly be considered conclusive.
In the seventh and final experiment, Williams tested the ease with which adult linguists could learn Esperanto and found that they could learn as much in six hours as a child can learn in four years. He uses this to encourage teachers around the world to learn Esperanto so that it can be implemented as an introductory language in all schools.
Paderborn, Germany; 1978
Perhaps the most famous study on the propedeutic value of Esperanto was conducted by Helmar Frank, a professor at the Institute of Cybernetics at the University of Paderborn. It is because of this study that the technique of teaching students Esperanto to prepare them for a natural language is often called “the Paderborn Method” on Wikipedia and Duolingo forums (a term that seems to have been coined by the MUIR bulletin). Its success spawned several followup studies by Günter Lobin, Sara Konnerth, and B. S. Meder, all researchers at the Paderborn Institute of Cybernetics, expanding on Frank’s results.
Unfortunately, I can find very little information about what any of them actually entailed. It seems that Frank’s original study was first written up in 1978 in his doctoral thesis, titled “Der propedeutische Wert von Plansprachen für den Fremdsprachenunterricht”. The MUIR bulletin cites the proceedings of a conference in Paderborn called “Laborkonferencoj: Interlingvistiko en Scienco kaj Klerigo”, which can apparently be obtained from the Institute of Cybernetics. However, I can’t find the original text of either. It seems that further results were published in Frank’s 1984 article “Zur Optimierung der Zeitverteilung bei Transferbewirkung durch Lehrstoffmodelle” in his own journal Grundlagenstudien aus Kybernetik und Geisteswissenschaft (GKG), but the only copies I’ve found are in libraries in Germany (I live in the US). That journal also seems to be where many of the followup Paderborn studies were published.
In 2005, Frank did provide to the newspaper Internacia Pedagogia Revuo an excerpt of a chapter from his book “Por dulingveco en Eŭropo: argumentoj kaj dokumentoj”. In that excerpt, he reports that students who had taken Esperanto for one year before starting English learned English 20% faster, and those who took Esperanto for two years learned English 35% faster. He also found that students who generally did poorly in school saw more benefit than those who generally excelled. An incomplete citation to Meder’s work in “Esperanto, lerneja eksperimento” by Helmut Sonnabend paints a similar picture.
It’s not clear to me from either of these whether the Esperanto replaced the first year of English study, or was an additional study period that the students in the control group did not have (if it’s the second one, then there’s not a direct comparison, so it seems premature to me to extrapolate that Esperanto would save time in the long run). They also don’t say how the control group was chosen, how the students’ speed of learning was assessed, or how many students were in each sample – three details that have proven problematic to many of the other studies on this topic. Such particulars are probably available in his thesis, which I really wish I could read. If any of ye have more details about the Paderborn experiment, please let me know. Until then, I think this is worth mentioning, but I remain skeptical of its conclusions.
In 1980, one Yukio Fukuda published the article “Zur rationalisierten Fremdsprach-Lehrplanung unter Berücksichtigung der (z. B. deutschen oder japanischen) Muttersprache”, in GKG. Unfortunately, Fukuda’s full text has proved just as elusive as Frank’s. According to a 1988 literature review in Studies in Second Language Acquisition written by Dan Maxwell, a system development official in Germany, Fukuda investigated whether Esperanto could help Japanese children learn English and found results similar to Frank’s. As with Frank’s work, I find it noteworthy, but am skeptical of its reliability without knowing what variables were controlled, what metrics were used, or what the sample size was. If any of ye know more about this, please tell me.
Genoa, Italy; 1989
In 1989, Elisabetta Formaggio, a teacher at the Rocca Elementary School in Italy, published the article “Lerneja eksperimento pri lernfacileco kaj transfero en la fremdlingvoinstruado” in the magazine Humankybernetik. I can’t find the text anywhere, but according to the MUIR bulletin, students in the experiment studied two years of Esperanto followed by three years of French, and were then evaluated on their French ability. It conspicuously omits the result. According to “Esperanto Fundamental”, a PDF I found on the internet by Roberto Figueira, a public official in Brazil, the result was a “transferência oculta K = 1.3.” I don’t know what that means.
Budapest, Hungary; 1995
In 1997, Katalin Smidéliusz, the administrator of edukado.net, published details from her doctoral dissertation as the booklet “Analisi Comparativa del Lessico Italiano Esperanto ed Ungherese a Fini Didattici”. It contains a systematic comparison of the Italian and Hungarian lexicons. Smidéliusz finds that 66% of Italian words are similar or identical to the corresponding Esperanto word and uses this, along with nonspecific reference to previous experiments, to support the claim that Esperanto has strong specific propedeutic value. I find the theoretical basis for this claim shaky, but in any case, this isn’t really an experiment, so it isn’t strictly relevant to the question at hand. I mention it only for the sake of completeness.
Melbourne, Australia; 1997
In 1997, Jennifer Bishop and Alan Bishop, both faculty in the education department at Monash University, published a report on the Monash University website. In it, they detail the EKPAROLI project, in which 21 students were taught either one year of Esperanto followed by two years of French, two years of Esperanto followed by one year of Japanese, or three years of Esperanto. These students compared with about 300 students from other schools who had been studying French, German, Indonesian, or Japanese for at least two years via questionnaires to the students and teachers regarding their attitudes toward the language(s) they had studied and their proficiency in their final foreign language. It was found that the Esperanto students ranked higher on average in both motivation and proficiency.
Even ignoring the small sample size and subjective measurements, I find this study more problematic than most. Given that the results are not broken up by the language they were being evaluated on, or by how long they had been studying it, it seems silly to me to even describe the non-Esperanto group as a control group. Some unspecified number of the Esperanto students were being tested on Esperanto, which is undoubtedly easier to learn than all of the languages the non-Esperanto students were tested on. The differences between the two groups were probably determined by how many of the Esperanto students were tested on Esperanto, and how many were tested on French or Japanese. The fact that the Bishops did not provide these numbers, even though doing so was surely within their power, and even though that information is critical to correctly interpreting these data, makes me very suspicious of their interpretation of their results.
Essex, UK; 2013
In 2013, Angela Tellier and Karen Roehr-Brackin, both faculty in the linguistics department at the University of Essex, wrote up a preliminary research report, which described an experiment on the effect of learning Esperanto on general linguistic awareness. In the same year, they published the booklet “Esperanto as a starter language for child second-language learners in the primary school”, and presented a paper on a related experiment at the annual conference of the European Second Language Association (EUROSLA), both on similar experiments. A final study was conducted later and published in the 2019 book “Learning Foreign Languages in Primary School: Research Insights”. All four were later summarized in a peer-review article published in Language Problems & Language Planning, a journal about language teaching.
The research report describes a small study in which 28 children aged 8 or 9 were tested for their general linguistic awareness, taught one year of either Esperanto or French, and then given the same test again. It found that both groups of students improved significantly, and that there was no significant difference in their general linguistic awarenesses either before or after the experiment. This is not surprising given the small sample size, and Tellier & Roehr-Brackin express a desire to repeat this experiment with more students and more time.
The booklet describes an experiment colloquially called “The Springboard to Languages Project”, in which a total of 358 children ages 7 through 11 were split up by age and taught Esperanto, sometimes in combination with other Western European languages, over the course of several years. Questionnaires and interviews were then conducted to determine the students’ attitudes toward and proficiencies with the studied languages, and their general awareness of linguistic structures. It found that Esperanto was enjoyable to the students and did increase their linguistic awareness. However, they were not compared to students who had not studied Esperanto, and Tellier & Roehr-Brackin consider this project encouraging but not scientifically conclusive.
The EUROSLA paper describes a more rigorous study in which 203 students were broken up into six groups and taught either Esperanto then Spanish, French then Spanish, French then Latin, French then Japanese, or just French. It is worth noting that Tellier & Roehr-Brackin did not assess students’ proficiencies in any particular natural language – only their general awareness of linguistic structures. Six of the eleven tasks used to test the students were based on a language that had been constructed specifically for the test, while the other five were based on different combinations of European languages.
They found that the Esperanto students performed about the same as the non-Esperanto students on all of the tasks except one: recognizing cognates among Dutch, English, Esperanto, Italian, Portuguese, Romanian, and Spanish (note that other tasks tested for recognition of cognates among other combinations of European languages). They posit that Esperanto could still have an effect that just wasn’t detected by the other ten tasks because studying a natural language after studying Esperanto erases all Esperantic bonuses except for the ability to recognize cognates in specifically Dutch, English, Esperanto, Italian, Portuguese, Romanian, and Spanish. That seems like grasping to me, though; it’s more likely that Esperanto had no special effect and one of the eleven tests just produced a significant result by chance, as often happens in frequentist studies.
They did also report that students who had studied Esperanto exhibited a smaller spread of scores than any of the other groups, implying that studying Esperanto provided an especially strong benefit for students who would have scored poorly, but provided less benefit than natural languages for students who would have scored well anyway.
The final study was the most telling. In it, 116 children ages 8 to 9 were taught either German, Italian, or Esperanto for one quarter, then French for one semester, then one semester of French. They were tested on their general linguistic awareness before and after the first semester, and then tested on their French proficiency after the second semester. The tests used were the same as for the previous studies. As with the EUROSLA study, the Esperanto group showed a levelling effect that the other groups did not. However, also as with the EUROSLA study, the students who had studied Esperanto did not perform significantly better than the students who had studied German or Italian, either on general linguistic awareness or French proficiency.
I consider the last two studies here to be by far the most reliable of the ones on this list, as they are the only ones that are well-documented, peer-reviewed, and based on a large sample size.
It is worth noting that there are several other studies cited in the MUIR bulletin, and in other places, that I have omitted here. For the most part, these are experiments that only tested whether Esperanto was easier to learn than natural languages, and did not test whether Esperanto helped with the learning of subsequent natural languages. While the relative ease of learning Esperanto is something of an open research question as well, I’m not interested in it personally because I don’t doubt that Esperanto is easier. Also, including those studies would make this literature review twice as long.
That is a decent number of experiments, but it is worth noting that most of them did not have adequate documentation, comparable control groups, large sample sizes, or objective measures of student success. It is also worth noting that there are still fairly substantial gaps. I haven’t found a single experiment testing the propedeutic value of Esperanto on adults, or the propedeutic value of Esperanto for studying non-European languages (the EKPAROLI project almost filled this gap, but we don’t know how many of their Esperanto students were actually evaluated in Japanese, so I’m not counting it). There also aren’t any studies that compare the propedeutic value of Esperanto to the propedeutic value of natural languages, except for Tellier’s and Roehr-Brackin’s studies, which only tested for one type of propedeutic value (and found that the natural languages were just as helpful as Esperanto was).
Even so, on the core question of “does Esperanto have strong propedeutic value, and if so, by which of the four mechanisms listed above?”, we can draw some conclusions.
Tellier & Roehr-Brackin are the only ones to have explicitly tested for general propedeutic value. Given that they found a null result in both their EUROSLA study and their 2017 study, which was also the two most reliable studies on this list, I think it’s safe to say that Esperanto does not have any more general propedeutic value than natural languages do. There’s certainly room for more studies, but with what has been done, it seems unlikely to me that they would turn up anything new.
For emotional propedeutic value, I think it’s quite another story. The three studies with the most detailed statistical analyses – Halloran, Frank, and Tellier & Roehr-Brackin – all observed a levelling effect in the Esperanto group; it helped along the students who normally would have struggled more than it did the students with natural aptitude, such that the class’s scores had a lower spread than the non-Esperanto group did. I posit that this is a signature of an emotional propedeutic effect. A student with aptitude for language learning will finish their first year feeling confident, no matter what foreign language they took; but for a student with less aptitude, whether their first foreign language is easy or hard will have a big impact on how they feel about language study in general. Going off of Halloran’s and Tellier’s & Roehr-Brackin’s numbers, it seems like this effect balances out on the whole such that a student of average intelligence who delays their French studies for a year to learn Esperanto will neither save nor lose any time in the long run, so whether the value is strong or weak depends on the individual.
Activity-derived propedeutic value value is trickier to pin down. Only one of the studies I found – Vilkki’s – was set up in a way that would detect activity-derived propedeutic value. Because he did not have a control Esperanto group that did not partake in those activities, it’s impossible to know whether the value he observed was activity-derived, specific, or both.
As for specific value, it’s hard to say. Because we can logically expect specific propedeutic value to show up most strongly for students learning a language that is very different from their own, Vilkki’s (Finnish students learning German), Fukuda’s (Japanese students learning English), and Smidéliusz’s (Hungarian students learning Italian) studies should be the best indicators (all of the others featured various combinations of English, French, and German, which are all relatively similar).[2,14] And since all three did find positive results, it would be reasonable to claim that Esperanto does have strong specific propedeutic value as a bridge between disparate languages. However, I think that the Esperanto activities in Vilkki’s study pose too big of a confusion for us to ascertain much, and I find it hard to trust Fukuda’s and Smidéliusz’s findings without seeing any of their numbers. So while the fact that these three studies all point toward the same result is promising, I think this is the aspect of propedeutic value that still most pressingly needs further research.
This raises the question, if Esperanto’s general propedeutic value is weak, as has been shown, and its emotional propedeutic value nets neutral on average, as the data suggest, and its activity-derived propedeutic value only appears in experiments that featured activities outside of class, as must be true by definition, and its specific propedeutic value is only apparent when the target language and native language are very different, as has been argued on theoretical grounds, then what caused the strong net benefit that Williams, Frank, Fukuda, and Bishop observed? I think the most likely explanation is that those studies were all simply flawed, either in their sample size, in their metrics of success, or in their statistical analyses. While I have stated my issues with Williams’s and Bishop’s studies, I don’t have any specific reason to think the other two did something wrong; I just find it hard to trust their results without reading them. Another possible explanation is that the emotional value is situationally dependent, and provided a net strong benefit to the students in these studies even though it didn’t in Eaton’s and Halloran’s. A third is that Esperanto has strong specific propedeutic value that is significant even between related languages like German and English. All I can say with confidence is that these data aren’t nearly as conclusive as many Esperantists make them out to be.
While the data are more compelling than I expected when I started this literature review, there is still more work to be done before anyone can truly claim to understand this phenomenon. It seems clear to me that teaching Esperanto before other languages equalizes students’ performance in the long run, but whether doing so is beneficial on average remains an open question. I think that Tellier’s & Roehr-Brackin’s 2013 work is an excellent example of what future experiments should aspire toward, and hope to see similarly quality research examining the other types of propedeutic value in the near future. Until then, though, perhaps Esperantists should refrain from advertising this feature too zealously. As Renato Corsetti said in 1997, “Oni ne promesu miraklojn, kiuj poste ne efektiviĝos.” One should not promise miracles that won’t come true.
And if you have more sources, please let me know in the comments, and I’ll update this document.
- Bishop, A. & Bishop, J.; “EKPAROLI Project report 1994 – 1997”; Retrieved from: angelfire.com/ok/andreo.
- Corsetti, R.; “Propede…kio?”; Internacia Pedagogia Revuo, 1997/4. ilei.info/pdf/Ipr141_rete.pdf/ipr/ipr_1997_4.htm
- “Diffusione del documento conclusivo della commissione per la promozione della lingua internazionale detta ‘Esperanto'”; Bollettino Ufficiale, Ministero della Pubblica Istruzione, Anno 122° (1995-06), N. 21-22; Retrieved from: parracomumangi.altervista.org.
- Eaton, H.; “Experiments in language learning”; The Modern Language Journal, vol. 19, no. 1 (1934-10), p. 1–4; doi.org/10.2307/315418
- “Esperanto as an international auxiliary language”; General secretariat of the League of Nations, 1922; Retrieved from: commons.m.wikimedia.org.
- Frank, H.; “Por dulingveco en Eŭropo: argumentoj kaj dokumentoj / Für Zweisprachigkeit in Europa. Argumente und Dokumente”; IFB Verlag & Akademia Libroservo, Paderborn (2005), p. 31–38; ISBN 3931263541.
- Frank, H.; “La Paderborna Eksperimento pri lingvo-orientiga instruado”; Internacia Pedagogia Revuo, 2014/1, p. 4–9; ISSN 1013-2031; Retrieved from: ilei.info.
- Halloran, J. H.; “A four year experiment in Esperanto as an introduction to French”; The British Journal of Educational Psychology, vol. 22, no. 3 (1952-09), p. 200–204; doi.org/10.1111/j.2044-8279.1952.tb02826.x
- von Hertzen, E.; “Joel Vilkki kaj Esperanto”; Esperanta Finnlando 1997/4, p. 12–13; esperanto.fi/eo/pri-ni/elehti/
- Maxwell, D.; “On the acquisition of Esperanto”; Studies in Second Language Acquisition, vol. 10, no. 1 (1988-02), p. 51–61; jstor.org/stable/44487440
- “Proposte per la conoscenza dell’Esperanto”; Città della Spezia; cittadellaspezia.com/2007/02/27/proposte-per-la-conoscenza-dellesperanto-19270/
- Setälä, V.; “Vizito al la eksperimenta lernejo en Somero, Finnlando”; Esperanto 1961/6, p. 107; uea.org/revuoj/esperanto/1961.
- Sonnabend, H.; “Esperanto: lerneja eksperimento”; kajeroj de Internacia Pedagogia Revuo; Edistudio, Pisa (1979); ISBN 88-7036-005-9.
- Smidéliusz, K.; “Analisi comparativa del lessico italiano esperanto ed ungherese a fini didattici”; COEDES, Milano (1997); ISBN 88-85872-09-3.
- Tellier, A. & Roehr-Brackin, K.; “The development of language learning aptitude and metalinguistic awareness in primary-school children: a classroom study”; Essex Research Reports in Linguistics, vol. 62, no. 1 (2013-04); Retrieved from: repository.essex.ac.uk.
- Tellier, A. & Roehr-Brackin, K.; “Metalinguistic awareness in children with differing language learning experience”; EUROSLA Yearbook, vol., 13, no. 1 (2013-01), p. 81–108; doi.org/10.1075/eurosla.13.06tel; Retrieved from: repository.essex.ac.uk.
- Tellier, A. (ed.); “Esperanto as a starter language for child second-language learners in the primary school” second edition; Great Britain: Esperanto UK (2013); ISBN 978-0-902756-35-9; doi.org/10.1075/lplp.37.1.12gre; Retrieved from: https://vdocuments.mx.
- Tellier, A. & Roehr-Brackin, K.; “Raising children’s metalinguistic awareness to enhance classroom second language learning; In: García Mayo, M. d. P. (ed.); “Learning foreign languages in primary school: research insights”; Bristol: Multilingual Matters (2017); ISBN 978-1-783098-10-1.
- Tellier, A. & Roehr-Brackin, K.; “Esperanto as a tool in classroom foreign language learning in England”; Language Problems and Language Planning, vol., 42, no. 1 (2018), p. 89–111; doi.org/10.1075/lplp.00013.roe; Retrieved from: repository.essex.ac.uk;
- Thorndike, E., Kennon, L., & Eaton, H.; “Language Learning: summary of a report to the International Auxiliary Language Association in the United States, incorporated, by Division of Psychology, Institute of Educational Research, Teachers College, Columbia University”, p. 27–30; Teachers College, Columbia University, New York City (1933); Retrieved from: worldcat.org.
- Vilkki, J. & Setälä, V.; “La eksperimenta instruado de Esperanto en la geknaba mezgrada lernejo de Somero (Suomio)”; El Popola Ĉinio, 1963/1, p. 38–39; Retrieved from: dlibra.kul.pl.
- Williams, N.; “A language teaching experiment”; The Canadian Modern Language Review, vol. 22, no. 1 (1965-12), p. 26–28; doi.org/10.3138/cmlr.22.1.26; Retrieved from: archive.org.
Edit: Added a section on the EKPAROLI Project.
Edit: Added a paragraph on Tellier’s & Roehr-Brackin’s first Esperanto study.
Edit: Added a clarification on the diversity of tests that Tellier & Roehr-Brackin performed.
Edit: Found and added the details of Smidéliusz’s work.
Edit: Deleted the section on Helmut Sonnabend, because it turns out that he only measured whether Esperanto was easier to learn than English.
Edit: Described Tellier’s & Roehr-Brackin’s fourth Esperanto study.