Forcorporate clients, the Amazon-owned service launched Audible for Business, which helps companies match skills-building audiobooks, workshops, and other spoken-word content to their individual employees and teams.
Please use the following spoken dictionary for assistance in pronouncing basic Manx words and phrases. The dictionary is organised according to themes and will be added to on a regular basis. Please let us know if there are particular terms you would like adding to the dictionary by contacting the Manx Language Development Officer for Culture Vannin: gaelg(at)
culturevannin.im
The Concise Dictionary of Spoken Chinese (1947), which was compiled by Yuen Ren Chao and Lien Sheng Yang, made numerous important lexicographic innovations. It was the first Chinese dictionary specifically for spoken Chinese words rather than for written Chinese characters, and one of the first to mark characters for being "free" or "bound" morphemes according to whether or not they can stand alone as a complete and independent utterance.
The influence of American structural linguistics, which shifted interest from the written to the spoken language, is evident in both the War Department's Dictionary of Spoken Chinese (1945) and Chao's and Yang's Concise Dictionary of Spoken Chinese (1947). "In both dictionaries we can observe the authors attempting not just to provide their Chinese entries with English equivalents but to demonstrate through grammatical categorization and examples how they are actually used".[7]
(1) The grammatical function of each word is distinguished according to whether it is free (F) or bound (B). With the exception of measure words or Chinese classifiers, called "auxiliary nouns" (AN), the dictionary generally did not indicate syntactic part of speech, called "word classes". Chao explains, "The same word, as a noun, means one thing; as an auxiliary noun (AN) something else, as a verb something else again. This is not a matter of inference, as those who say that Chinese has no parts of speech assume, but a matter of individual facts.".[14] The dictionary's English translation equivalents usually can clarify Chinese part of speech; if 吃 chy [chī] is defined by the English verb "to eat", then it is itself also a verb. word classes are only specified in cases of ambiguity; 脂肪 jyfang [zhīfng] "fat" is marked n. "noun" since English "fat" can also be an adjective.[15] The dictionary lists other specialized grammatical categories,[16] for instance, "auxiliary nouns proper" and "quasi-auxiliary nouns", and introduces for the first time in a Chinese dictionary "many new ideas about the linguistic structure of Chinese, such as the four types of verbal complements": the "pre-transitive," "verb-object construction", "possessive object," and "impersonal verb-object compound"".[17]
(8) All entries "are treated as morphemes, or monosyllabic meaningful spoken words", whether bound morphemes or free morphemes, rather than as characters. Chao gives a chemical analogy to differentiate between zdiǎn 字典 "character dictionaries" and cdiǎn 辭典 "word dictionaries".
In China, dictionaries are divided into 字典 and 詞典, the former giving only single characters, which may be compared with chemical elements, and the latter compounds and phrases, like chemical compounds. To pursue the chemical analogy one step further, compounds are so numerous that they cannot all be included except in a much more comprehensive work. We can do more than merely list the elements and their atomic weights. We can classify their affinities, their electric polarity, indicate whether they can be ionized, and give such information as to enable us to predict more compounds than can be listed. Thus by giving the morphemes of the language properly analyzed, indicating whether they are bound or free, the attempt has been made to give the equivalent of a dictionary of compounds within the space of a dictionary of single words."[21]
By treating all entries as bound or free morphemes rather than as characters, Chao and Yang have made an attempt to give the equivalent of a dictionary of compounds within the space of a dictionary of single words.[20]
Most reviewers have praised the Concise Dictionary of Spoken Chinese, while some have been critical. The co-author Lien-Sheng Yang responded to DeFrancis' and Simon's reviews in a 1949 article about free and bound morphemes in Chinese.[24]
The Chinese linguist Luo Changpei describes the dictionary as "unprecedented in the history of Chinese-European lexicography since its beginnings" in the early 17th century.[25] Luo lists three unique features of the dictionary, combining six of the eight given by Chao (above); the first combines (1) and (3), the second (2), (7), (8), and the third is (6). Luo[26] lists 15 corrections or suggestions, 9 of which are included in later editions of the dictionary, under Corrections and Additions.[27]
DeFrancis suggests that Chao and Yang have been "unduly influenced by the ideographs and the myths of Chinese monosyllabism".[28] Although the dictionary title has "Spoken Chinese", the authors approach the subject through characters rather than through speech. Using data from a 10-page sample, DeFrancis disproves Chao's contention that with "very few exceptions" Chinese morphemes are "for the most part monosyllabic".[21] The sample dictionary entries marked as "literary" (L), "comparable to yclept in English and hence not really belonging in a dictionary of spoken Chinese", amount to 16% of the total entries. (Yclept is an archaic or humorous word meaning "called; named".
Of the entries which really represent spoken forms, no more than 29 per cent have been classified by the authors as Free, the only category which is generally accepted as designating a word in English and other languages. But not all the remaining 71 per cent are classified as Bound forms of the type er in banker or sender. Only 49 per cent are of this type of meaningful syllables without independent life. The remaining 22 per cent, represented by shan and hu in shanhu, "coral," are of a type which have no more meaning or independence than do cor and al in the English equivalent.[29]
DeFrancis proposes that the category of Bound syllables (not "words") should be divided into two groups: "meaningless bound syllables" (like shan and hu in shanhu 珊瑚 "coral") and "meaningful bound syllables" (like fu "father" and mu "mother" in fumu 父母 "parents"). "The former are fully bound (occurring in only one word), and the latter are semibound (occurring in more than one word)." DeFrancis concludes that the contributions of scholars like Dr. Chao, "give hope that lexicographers, if they can concentrate on Chinese speech and not be misled by the ideographic writing, will eventually succeed in compiling a real dictionary of the Chinese spoken language".[29]
In response to DeFrancis' review, Lien-sheng Yang states that comparing the dictionary entries designated as literary to yclept in English is "misleading, because the latter is an archaic word, whereas the former are still used in modern Spoken Chinese".[30] Yang says DeFrancis' suggestion of differentiating "meaningless bound syllables" and "meaningful bound syllables" appears interesting but unfortunately it involves three difficulties. First, since the word meaning is ambiguous, a linguist has to define "meaningful" and "meaningless" and ascertain whether all native-speaker informants agree. Second, a linguist needs to consider differences in the informants' background and education, "One syllable which is meaningful to one may be meaningless to another".[30] Third, the identification of meaningful and meaningless syllables with those occurring in more than one word and those occurring in only one "is doubtful". Taking the example of shan and hu in shanhu "coral", Yang notes both characters are used in other compounds, namely, shanshan 珊珊 "tinkling sound (of ornaments)" and hulian 瑚璉 "two types of ritual vessels".[31]
The German sinologist Walter Simon says the Concise Dictionary of Spoken Chinese is a "definite advance on our knowledge of the Chinese language" and calls it a "very important lexicographical contribution" from which "students cannot fail to derive great benefit".[19] Simon says
The derivation method for this variable is specified within the Official Languages (Communications with and Services to the Public) Regulations, registered on December 16, 1991, in accordance with section 85 of the Official Languages Act, R.S.C., c. 32 (4th Suppl.).
The method used to derive the first official language spoken variable takes into account, first, the knowledge of official languages, second, the mother tongue, and third, the home language. A complete description of this method is available on the Definitions, data sources and methods page for the language variables on Statistics Canada's website.
For more information on language variables, including information on their classifications, the questions from which they are derived, data quality and their comparability with other sources of data, please refer to the Languages Reference Guide, Census of Population, 2016.
Canada owes the success of its statistical system to a long-standing partnership between Statistics Canada, the citizens of Canada, its businesses, governments and other institutions. Accurate and timely statistical information could not be produced without their continued co-operation and goodwill.
Statistics Canada is committed to serving its clients in a prompt, reliable and courteous manner. To this end, the Agency has developed standards of service which its employees observe in serving its clients.
Mah Meri is an Aslian (Austroasiatic: Mon-Khmer) language spoken in scattered settlements along a section of the southwest coast of Selangor in Peninsular Malaysia. The Mah Meri are the only Aslian speakers who live in a coastal environment. Their language, which may have about 2,000 speakers, has no written language and is highly endangered.
3a8082e126