Editing Cardinal vowels (section)

==Limits on the accuracy of the system==
The usual explanation of the cardinal vowel system implies that the competent user can reliably distinguish between sixteen Primary and Secondary vowels plus a small number of central vowels. The provision of diacritics by the [[International Phonetic Association]] further implies that intermediate values may also be reliably recognized, so that a phonetician might be able to produce and recognize not only a [[close-mid front unrounded vowel]] {{IPA|[e]}} and an [[open-mid front unrounded vowel]] {{IPA|[ɛ]}} but also a [[mid front unrounded vowel]] {{IPA|[e̞]}}, a centralized mid front unrounded vowel {{IPA|[ë]}}, and so on. This suggests a range of vowels nearer to forty or fifty than to twenty in number. Empirical evidence for this ability in trained phoneticians is hard to come by.

Ladefoged, in a series of pioneering experiments published in the 1950s and 60s, studied how trained phoneticians coped with the vowels of a dialect of [[Scottish Gaelic]]. He asked eighteen phoneticians to listen to a recording of ten words spoken by a native speaker of Gaelic and to place the vowels on a cardinal vowel quadrilateral. He then studied the degree of agreement or disagreement among the phoneticians. Ladefoged himself drew attention to the fact that the phoneticians who were trained in the British tradition established by Daniel Jones were closer to each other in their judgments than those who had not had this training. However, the most striking result is the great divergence of judgments among ''all'' the listeners regarding vowels that were distant from Cardinal values.<ref>{{cite book|last1=Ladefoged|first1=P.|title=Three Areas of Experimental Phonetics|date=1967|publisher=Oxford University Press|pages=132–142}} See especially Figure 47 on p.&nbsp;135</ref>