Orthographic reform and psycholinguistics: a selective synthesizing review
Emil Kirkegaard
Introduction
Linguistics is by now an established field. As always happens when fields grow, subfields form and researchers from different subfields tend to not collaborate ‘across borders’. In this paper I attempt to synthesize research from orthographology (the study of orthographies) and psycholinguistics.
Language features and ease of use and learning
Known human languages differ in almost every imaginable way. Some have many cases; some have only one. Some are polysynthetic; others are strongly analytic. And some have spellings where there is a clear correspondence between graphemes and phonemes, while others don’t. Though arguably many languages’ features have a potential effect on ease of use and acquisition (e.g.morpheme to word ratio, use of affixes, use of agreement), I will focus on orthography in this paper.
Orthography
One may define orthography along a continuum of narrow and broad definitions. Most narrowly, the orthography of a language refers simply to the standard spellings codified in a dictionary published by an official language board, or a similarly anointed work.More broadly, one may define it to include the choice of the script, rules governing such items as punctuation, Capitalization, abbreviation, formatting(underlined, bold, italic), apostrophes, exclamation/question marks, and so on, including both official and colloquial usage, a significant disparity being found often between the latter and the former. This article is primarily concerned with the spelling of words, instead of the rather ancillary details of broadly defined orthography.
Orthography changes and reforms
Some languages have no central language academy which has the (legal) authority to decide what the ‘correct’ orthography should be. Orthographies in these languagesmorph gradually and continually as usersintroduce small variations into written language. This is the case for the world’s lingua franca, English.
However, languages which have a central language board can initiate top-down (rather than bottom-up) changes, enabling large, fast-acting transformations often called language reforms. While English has not had major orthographic reforms lately, other European languages have. Some have been minor, such as the bitter“Comma War” (“kommakrigen”) for Danish, which involved a single rule concerning comma placement (see e.g. Davidsen-Nielsen, 2009).
Another very bitter one was the German spelling reform of 1996, which, among other things, changed the way <ß>/<ss> is used. The reform even sparked a legal case going all the way to the supreme court in 1998 (Federal Constitutional Court of Germany) about who has the authority to devise the spellings(Johnson (2005)).
The above two cases are relatively minor reforms. The Turkish reform of 1928 was thorough, changing the script as well as replacing a large number of foreign words with neologisms (“purification”). Before the reform, Turkish used an Arabic script ill-suited to represent the rich vowel system of Turkish (Dogançay-Aktuna, 2004).
Reasons for reforming
Given the problems with implementing language reforms in general, why would anyone bother?
The first reason, and perhaps the most common, is nationalistic sentiment. This was clearly the case in Turkey and is also the case in the many European countries that have official language institutes or boards (e.g. France, Iceland). Usually the aim of these boards is to prevent ‘deterioration’ of the language, usually by coining new words for loan words (which attempts to prevent loan words from becoming the standard word).
Many linguists are averse to this kind of activism which usually goes by the name “linguistic prescriptivism” (Fromkin, 2011:14ff, McGregor, 2009:2ff). Their aversion is understandable. The talk of ‘deteriorating’ language reminds one of 20th century racists/racialists warning about the dangers of cross-racial mating (or “miscegenation”).
The second reason why some (modern) linguists and others are adverse to prescriptive concerns is that they seem contrary to the natural course of languages. Just like species, languages ‘evolve’. In this line of thinking, to ‘fiddle’ with languageretards natural processes and is therefore undesirable (Dogançay-Aktuna, 2004). This kind of reasoning, however, is a logical fallacy known as the naturalistic fallacy or appeal to nature (Curtis, 2014).
Good reasons for reforming?
The second reason for reform, insisted upon only by a small number of people and societies such asThe English Spelling Society ( and the Simplfied Spelling Board [1906-1920] (Wikipedia, 2014) , is that reform is necessary every once in a while. This is because, while the spoken language will change with no artificial prompting, sometimes quickly (e.g. English and Danish vowel changes), and is also difficult to direct, written language is a different case. It has been kept rigid by the introduction of official orthographies and dictionaries around the time of the introduction of the printing press (see e.g. Bell, 2004). Prior to this, writing was more in line with local pronunciation and thus varied and changed in tandem with spoken language.
Rigid spellings cause the connection between the phonemes and the graphemes (which may have been poor to begin with) to deteriorate, resulting in multiple ways to: a) pronounce a given grapheme, and b) write a given grapheme. Spelling reform advocates argue that this affects how easy it is to learn a language, and given the especial importance of mastering language in today’s information society,that reforming the spellings to fit spoken language is a very important social goal.
They argue furthermore that the irregular and inconsistent spelling system of English is a major cause of low literacy levels, and that it wastes crucial time in the educational system. This time could be better spent learning something more useful than the arbitrary spellings of words.
Reformists generally argue their case by showcasing the many irregular spellings of English. In one useful overview, Masha Bell (2009, 2011) writes that English has 43.5 ‘sounds’ (phonemes?) but has 185 spellings for these. Furthermore, she writes that there are 91 ‘main’ ways to spell, but only 11 of them without exceptions, namely:
bed, jug, gorge, ring, single, pin, musician, this, thing, vantelevision.
[b] [dʒ] [dʒ] [ŋ] [əl] [p] [ʃən] [ð] [θ] [v] (phonemes are mine)
She then goes on to give examples of same that can be spelled in more than one way. For instance the dipfthong [eɪ]:
plate - wait weight straight great vein reign table dahlia champagne fete
Similar illustrative arguments are given by other reformists (Webster, 1789; Upward, 1992; Ripman and Archer, 1948; Cleckler, 2005; Kirkegaard, 2014).
In general, the literature on language reform does not contain references to experimental, empirical literature that support their claims, so they remain plausible but unproven assertions. This is unfortunate because there is a rich literature in linguistics on matters of spelling. Below is a review of the psycholinguistic experimental literature on (ir)regular and (in)consistent spellings.
The psycholinguistic literature and spellings
There are various ways to find out if a particular word is comparatively hard to read. One method is to measure the time it takes for people to read it by using equipment that follows the gaze of readers (see for introduction Rayner and Pollatsek, 2006; Balota et al, 2006; Traxler, 2011:Chapter 10). A second method is to have people read words out loud and measure both how long it takes and the percent of ‘errors’. Both have been used to study the effects of orthographies.
One way to think about the relationship between phonemes and graphemes is to look at the word level and ask two questions about word pairs: 1) do they have the same spelling?2) do they have the same sound? This gives four kinds of pairs:
Same sound / Not same soundSame spelling / Homophonic homographs / Heterophonic homographs
Not same spelling / Homophonic heterographs / Heterophonic heterographs
Reformers prefer spellings that make homophones that are not homographs become so, e.g. bean, beenbeen.[1] Some critics of language reform claim that proposed reforms willincrease confusion for readers and writers by increasing the percent of homographs. This claim is based on the theory that the reader is better able to distinguish between words if they have different spellings even if they have the same sound. This is a testable claim since it implies that readers should have less trouble reading homophonic heterographs than homophonic homographs. However, this is not what Folk and Morris (1995) found. There seemed to be little difference (Tables 1, 2, 5) in the reading times of the relevant pairs.
Reformers also want heterophonic words to be heterographic. This is based on the model that reading words that have more than one pronunciation associated with them slows the reader down, which is in line with what Folk and Morris (1995) found. When readers make incorrect assumptions about the pronunciation of a particular word and must correct afterward, they move their focus back to the ambiguous word for reprocessing (called regressions). Regressions were not higher for either homophonic heterographs nor homophonic homographs, but they were higher for heterophonic homographs as might be expected (Table 4).
Furthermore, results showed that readers were slower to name words heterophonic homographs (Table 9) and made more errors in their pronunciation. Also, there was no response time advantage in homophonic heterographs over controls, which was predicted by the model that having different spellings for different words helps readers, and readers did not make fewer pronunciation errors with these either (Table 10). In fact, they made more errors, but the small sample sizes preclude confidence in these differences. Error rates for the pronunciation of heterophonic homographs was very high compared to control (Tables 10, 12).
Similarly,Van Orden (1987) used a category test and found similar results, in which subjects are shown a word and must decide as quickly as possible whether it belongs to a specific category. E.g. the subject is shown the word <DUCK> and asked whether it belongs to the category carnivores. The experimental trick is to show subjects words that have a homophonic pair which belongs to the category, while the word itself does not. For instance, rows does not belong to the category ‘flowers’, but rose does. Results show that when readers are shown a word which has a homophonic pair word which has another relationship to the category than the target word, the error rate goes up.
Van Orden et al. (1988) replicated this result for nonword homographs. Even reading a nonword like sute tends to make people answer yes to the question ‘Is it an article of clothing?’.
Seidenberg et al. (1984) found that subjects were slower to name low frequency words with irregular and/or inconsistent spellings and responded slower to lexical decisions. This seems to show that having a irregular/inconsistentorthographydoes not just mean that people learn to read and write later, but that even adult readers fare worse with an orthography like that.
Lesch and Pollatsek (1993), Perfetti and Bell (1991), Perfetti et al. (1988) found evidence that reading proceeds initially through the phonological relationship (i.e. through grapheme to phoneme to morpheme to meaning). This means that spelling homophones differently will not remove the association as is sometimes suggested by critics. This fits well with the above discussed experiments.
When reading, the focus of the eyes (fovea) is on the target word. However, extra information is also gained from the next word near the focused area (parafovea). Pollatsek et al. (1992) showed that homophones when viewed in the n+1 position sped up the identification of the target word, showing further that it is sound that is basic, not the visual. See also Rayner et al. (1995) for a similar study.
Jared et al. (1990) found that subjects were slower to name words with inconsistent spellings. Further, they found that subjects were also slower with words which belong to an orthographic neighborhood with members with divergent spellings. This can be seen as a reason to avoid heterophonic homographs. E.g. to change the spelling of <pint> to <pient> so that there is no longer an inconsistent neighborhood consisting of <mint>, <tint>, and <pint>.
Glushko (1979) found that students were slower to read aloud words with irregular spellings such as have (compare with save). There was also a neighborhood effect as in the Jared et al (1990) study.
English is the not only language with an inconsistent and irregular orthography. However, most research has been done on this language because it is the major spoken language of linguists. It is a silver lining that English happens to have a very (probably the most) irregular and inconsistent spelling of all languages. If the lingua franca happened to be Esperanto or Finnish, than we would have perhaps not discovered these effects.
Interestingly, similar results have been found for Chinese(Mandarin, Cantonese), a very different language than English sincetheyuse an ideographic script. Tsai et al. (2005) found that subjects were slower to name inconsistent signs than controls. The effect was only found for infrequent words, a result found for English too. See also Hsu et al. (2009), Lee et al. (2007).
Future research
Future studies should replicate the prior experiments for English with larger samples, as well as languages other than English. Most European languages are irregular to some degree, so there are enough languages to study. Especially suitable are Danish and French as these are known to have irregular and inconsistent spellings.
References
Balota, D. A., Yap, M. J., Cortese, M. J. (2006). Visual Word Recognition: The Journey from Features to Measing (A Travel Update). In Traxler, M. J., Gernsbacher, M. A., & Cortese, M. J. (2006). Handbook of Psycholinguistics. Academic Press.
Bell, Mashra. (2004). Understanding English Spelling. Pegasus Elliot Mackenzie Publishers.
Bell, Masha. (2009). Rules and Exceptions of English Spelling. Pegasus Elliot Mackenzie Publishers.
Bell, Masha. (2011). Why is English spelling exceptionally irregular? The English Spellilng Society.
Cleckler, B. C. (2005). Let's End Our Literacy Crisis: The Desperately Needed Idea Whose Time Has Come. American Book Publishing.
Curtis, G. N. (2014). Appeal to nature. In FallacyFiles.
Davidsen-Nielsen, N. (2009). Moders stemme, fars hammer: en debatbog om dansk sprogpolitik. Dansklærerforeningens forlag.
Dogançay-Aktuna, S. (2004). Language planning in Turkey: yesterday and today. International journal of the sociology of language, (165), 5-32.
Folk, J. R., & Morris, R. K. (1995). Multiple lexical codes in reading: Evidence from eye movements, naming time, and oral reading. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(6), 1412.
Fromkin, V., Rodman, R., & Hyams, N. M. (2011). An introduction to language. 3rd ed. Cengage Learning.
Glushko, R. J. (1979). The organization and activation of orthographic knowledge in reading aloud. Journal of Experimental Psychology: Human Perception and Performance, 5,674–691.
Hsu, C. H., Tsai, J. L., Lee, C. Y., & Tzeng, O. J. L. (2009). Orthographic combinability and phonological consistency effects in reading Chinese phonograms: an event-related potential study. Brain and Language, 108(1), 56-66.
Jared, D., McRae, K., & Seidenberg, M. S. (1990). The basis of consistency effects in word naming. Journal of Memory and Language, 29,687–715.
Johnson, S. A. (2005). Spelling Trouble?: Language, Ideology and the Reform of German Orthography. Multilingual Matters.
Kirkegaard, E. (2014). Lyddansk: Et reformforslag til dansk retskrivning.
Lee, C. Y., Tsai, J. L., Chan, W. H., Hsu, C. H., Hung, D. L., & Tzeng, O. J. (2007). Temporal dynamics of the consistency effect in reading Chinese: an event-related potentials study. Neuroreport, 18(2), 147-151.
Lesch, M. F., & Pollatsek, A. (1993). Automatic access of semantic information by phonological codes in visual word recognition. Journal of Experimental Psychology: Learning, Memory, and Cognition, 19(2), 285.
McGregor, W. B. (2009). Linguistics: An Introduction. Continuum.
Perfetti, C. A., Bell, L. C., & Delaney, S. M. (1988). Automatic (prelexical) phonetic activation in silent word reading: Evidence from backward masking. Journal of Memory and Language, 27(1), 59-70.
Perfetti, C. A., & Bell, L. (1991). Phonemic activation during the first 40 ms of word identification: Evidence from backward masking and priming. Journal of Memory and Language, 30(4), 473-485.
Pollatsek, A., Lesch, M., Morris, R. K., & Rayner, K. (1992). Phonological codes are used in integrating information across saccades in word identification and reading. Journal of Experimental Psychology: Human Perception and Performance, 18(1), 148.
Rayner, K., Sereno, S. C., Lesch, M. F., & Pollatsek, A. (1995). Phonological codes are automatically activated during reading: Evidence from an eye movement priming paradigm. Psychological Science, 6(1), 26-32.
Rayner, K. and Pollatsek, A. (2006). Eye-Movement Control in Reading. In Traxler, M. J., Gernsbacher, M. A., & Cortese, M. J. (2006). Handbook of Psycholinguistics. Academic Press.
Seidenberg, M. S., Waters, G. S., Barnes, M. A., & Tanenhaus, M. K. (1984). When does irregular spelling or pronunciation influence word recognition?. Journal of Verbal Learning and Verbal Behavior, 23(3), 383-404.
Traxler, M. J. (2011). Introduction to psycholinguistics: Understanding language science. John Wiley & Sons.
Tsai, J. L., Su, E. C. I., Tzeng12, O. J., & Hung12, D. L. (2005). Consistency, regularity, and frequency effects in naming Chinese characters.
Upward, Christopher. (1992). Cut Spelling: A handbook to the simplification of written english by omission of redundant letters. Simplified Spelling Society, 1992.
Van Orden, G. C. (1987). A ROWS is a ROSE: Spelling, sound, and reading. Memory & Cognition, 15(3), 181-198.
Van Orden, G. C., Johnston, J. C., & Hale, B. L. (1988). Word identification in reading proceeds from spelling to sound to meaning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 14(3), 371.
Ripman, Walter; Archer, William; Jones, Daniel (ed.) and Orton, Harold (ed.). (1948). New Spelling. Being proposals for simplifying the spelling of English without the introduction of new letters. 6th edition. Sir Isaac Pitman & Sons, London.
Webster, Noah. (1789). "An Essay on the Necessity, Advantages, and Practicality of Reforming the Mode of Spelling and of Rendering the Orthography of Words Correspondent to Pronunciation," Dissertations on the English Language: With Notes, Historical and Critical, to Which is Added, by Way of Appendix, an Essay on a Reformed Mode of Spelling, with Dr. Franklin’s Arguments on That Subject. Boston. pp. 391. 393-98. 405-6.
Wikipedia. (2014). Simplified Spelling Board. (2014, April 21). In Wikipedia, The Free Encyclopedia. Retrieved 11:52, June 10, 2014, from
[1] A useful list of homophonic heterographs can be found here: