0. Introduction. In North America--and particularly in the United States--minority languages generally die out rather quickly to make way for the surrounding language. Certain pockets of immigrant languages, however, can still be found; one of the best known of these is the language that has become known variously as Pennsylvania Dutch, Pennsylfawnisch, and Pennsylvania German (the preferred term in the scholarly literature, hereafter abbreviated PG), a dialect of German which has been spoken continuously in North America1 since the end of the seventeenth century by a group of people which have come to be known as the Pennsylvania Germans.2
PG is spoken by various groups of people, primarily by Anabaptist "plain people"--that is, the conservative Amish, Mennonite, and Hutterite groups. (While the language was spoken by other groups of German origin as recently as fifty years ago, it is only rarely heard now among those populations.) The area in which the language is spoken covers a roughly diamond-shaped area with corners in southern Ontario, southeastern Pennsylvania, southern Maryland, and at the Indiana-Illinois border, and throughout the area PG is in some degree of contact with North American English. This has given rise to what is generally described as a stable bilingual situation, with all or nearly all speakers of PG being fluent in North American English; therefore, studies of PG speakers do not have to be limited to PG, but can also look at the English spoken by PG speakers, comparative fluency in the two languages, etc.
This paper will attempt to give an overview of the research which has been done on the languages of the Pennsylvania Germans over the past 125 years. To accomplish this end, the paper will look at the evolution of the field of study, starting with the studies of PG made in the late nineteenth century, followed by the dialectological surveys of PG and the Pennsylvania Germans done in the first half of the twentieth century as reflected primarily in articles published in American speech at that time, followed by studies done since the time of the 1968 Tenth Germanic Languages Symposium at the University of Texas at Austin, which focused in great part on the German languages spoken in North America and how to maintain their presence. These divisions will allow a glimpse of the way in which the study of the languages of the Pennsylvania Germans has developed, from a focus on structural description to dialectology to language death and language maintenance; they also allow highlights of the field and its development to be presented, as the numbers of studies which have been published is much larger than the number that could even begin to fit into a paper of this length.
1. The nineteenth century. The nineteenth century saw the end of the Anabaptist migration to North America (in about 1860), and the first widespread realization that a large group of people in North America were maintaining their German language. The first records we have of this are occasional entries in the travel journals of German-speaking tourists and occasional general-interest books published about PG (Yoder 1971); these are, however, not very important for present purposes, whereas the scholarly literature on PG that began to appear in the last quarter of the nineteenth century is.
The two main scholarly works on PG to come out of the nineteenth century are Haldeman's Pennsylvania Dutch: A dialect of south German with an infusion of English (1872) and Learned's The Pennsylvania German dialect (1889). These works are now most useful because they offer a snapshot of the condition of PG in the late nineteenth century; they also give no hint that non-Anabaptist PG speakers were in any danger of losing their proficiency in PG. It thus seems that that started to happen sometime in the late nineteenth or early twentieth centuries, as dialectologists were already reporting on that phenomenon in the 1930s.
2. Twentieth century dialectologists. PG and the area populated by PG-speaking people have been of great interest to dialectologists. In the first half of the twentieth century this interest focused on two issues, each of which will be dealt with separately--the structure of PG itself, and the English spoken by Pennsylvania Germans as well as non-Pennsylvania Germans living in areas with large numbers of PG speakers. As the question of PG itself received more attention, it will be dealt with first.
2.1. Twentieth century studies of PG. The study of PG by dialectologists began to appear in the scholarly press in the 1920s with books and articles by various authors, most notably Lambert (1924), who worked on developing dictionaries of PG. Other PG dictionaries and grammars followed; the field of study, however, flowered particularly brightly in the pages of American speech around the time of the Great Depression and World War II. The first article in American speech on PG (Follin 1929) identified PG as a "southern German" dialect, echoing the conclusion that Haldeman (1872) had reached half a century earlier. A quick response came from Bickel (1930), who noted that Follin (1929) had identified PG as a Palatinate German dialect. Bickel also noted that the use of PG was fading, but it retained a strong hold in certain counties of Pennsylvania;3 this may be a reflection of the shift of the non-Anabaptist Pennsylvania Germans away from speaking PG and toward speaking English in areas where the PG-speaking populations were not concentrated enough to hold out against the encroachment of English.
A favorite study of dialectologists studying PG has been the borrowing of English words into PG. Haldeman (1872) and Learned (1889) had both made note of large numbers of English borrowings, but comprehensive dialectological studies of these borrowing were not done until the issue was taken up by such people as Werner (1931), Buffington (1941), Frey (1942), Reed (1942,1948), and Schach (1948,1949,1952,1954); in addition, Schach (1951) looked at semantic borrowing from English into PG. These studies often made it seem that the German of PG was simply being replaced by English words, but some researchers pointed out that what had happened was only that certain high-frequency words had been borrowed, giving the appearance that PG was being overwhelmed by American English, when in fact most of the words in PG are clearly German in origin.
Studies of PG from this time period that focused on its character as a German language were rarer, but still existed; representative of these are the studies by Buffington (1937,1939,1948) and Frey (1941,1943). These sorts of studies were important for various reasons, but two of them are worthy of particular note. In his study of the diphthong oi, Frey (1943) demonstrated that the differences between PG and European German were actually the result of regular historical processes much like, for example, those differentiating northern and southern varieties of European German. Buffington's (1939) study of Pennsylvania German in relation to other German dialects also focused on historical processes, but looked at which dialect of European German PG is most closely related to; Buffington's finding was that PG is most closely related to Palatinate German (as had already been concluded by several researchers), and that it is most closely related to eastern Palatinate rather than western Palatinate. This came as a surprise, as most of the early German settlers in the present PG area came from the western Palatinate region; however, Buffington's evidence is most compelling, and provided a useful tool for further historical studies of PG and its development.
2.2. Twentieth-century studies of PG-influenced English. Some of the dialectologists working at this time chose to focus on the English spoken in PG-speaking areas; these studies include those done by Struble (1935), Page (1937), Frey (1945), Wilson (1948), and Ashcom (1953). As each of these studies is important for a different reason, the following paragraphs will deal quickly with each of them in turn.
2.2.1. Struble (1935). Struble's (1935) report of the English of PG speakers concluded that speakers of PG speak English with a particular accent, affirming the conventional wisdom of both that time and the present. Struble listed various features of PG-accented English (such as exchanging w and v, devoicing of English j, etc.), and claimed further that these features were stable, or in other words that one could expect all Pennsylvania Germans to exhibit them when speaking English.
2.2.2. Page (1937). Page's (1937) study of the English of PG speakers built on Struble's (1935) study. Page noted that Struble had looked at a variety of PG-influenced English that was very divergent from Standard American English; Page looked at PG-influenced English varieties which more closely approached the standard.4 This study is most valuable for two reasons: first because it pointed out that there was a continuum (or at least a multi-tiered hierarchy) of levels of PG influence on the English of PG speakers, and second because it pointed out the inverse relationship between literacy in English and the use of PG-influenced forms in English.
2.2.3. Frey (1945). Frey (1945), in one of the most-referenced articles on the subject of PG-influenced English, looked at the languages spoken by the Old Order Amish (OOA) and concluded that the OOA of Lancaster County, Pennsylvania are trilingual (in PG, a German dialect more closely approaching written European German called Amish High German, and PG-influenced English). This remains a controversial conclusion, which some, such as Meister Ferré (1991), reject, concluding that even if Frey was correct about the situation in 1945, it was only ever the case in Lancaster County and is no longer the case (that is, that the OOA are now balanced PG-American English bilinguals). Others, such as the members of the Essen-Delaware Amish Project Team (Enninger et al 1984), accept Frey's conclusions and hold that the OOA remain trilingual.5
2.2.4. Wilson (1948). Wilson (1948) conducted a fairly cursory study of words used by PG speakers when speaking English, and concluded that much of what strikes speakers of other varieties of American English as odd about PG-influenced English stems from an "attempt to speak English as a literal translation of the German idiom" (p. 236). Unfortunately, all that is then given is a description of particular items, without any mention of how often and when they were used or what sort of support exists for the claim that PG interference is the result of such "literal translation."
2.2.5. Ashcom (1953). Ashcom's (1953) study of the English of central Pennsylvania does not deal directly with any of the languages of the Pennsylvania Germans, but it is significant in that it points out that PG has had an influence on the English of certain regions even among non-Pennsylvania Germans. Ashcom included with the article a short list of lexical variants found among all speakers in the PG-speaking region; among these are PG-influenced items such as smearcase 'cottage cheese' and simple regionalisms such as onion snow 'a late, short-lived snow in the spring.'
3. Recent studies. A linguistic atlas of Pennsylvania German (Reed and Seifert) was published in 1954 as the culmination of the dialectological work which had been done in the preceding years in the PG-speaking region of eastern Pennsylvania. After this point, interestingly, dialectological studies of the Pennsylvania Germans were still occasionally conducted (see, for example, Seifert 1971 and Shields 1985), but the focus on PG and the languages of the Pennsylvania Germans seems to have trailed off among dialectologists. Despite the gap in study of PG--or possibly because of it (see Gilbert 1971b)--the Tenth Germanic Languages Symposium at the University of Texas at Austin dealt to a great extent with the status of long-term communities of German speakers in the United States. The years that followed saw a new resurgence in studies of the languages of the Pennsylvania Germans, with most of the researchers concentrated in the mid-Atlantic and Great Lakes regions of the United States, as well as Essen, Germany. The remainder of this paper will be organized somewhat like the section on studies conducted in the first half of the twentieth century--it will deal first with studies of PG itself, and then with studies of English as spoken by the Pennsylvania Germans.
3.1. Recent studies of PG. As the twentieth century has continued, linguistic and cultural pressure on the remaining PG speakers has increased (see, among others, Hostetler 1993; Kraybill 1994). Reflecting this, most recent studies of PG have tended to deal with issues of language death, language maintenance, and linguistic pressure from non-PG populations. Although it is difficult to draw lines between these sorts of studies, the following sections will deal with studies of communities in which PG is dropping out of use separately from the other studies.
3.1.1. Recent studies of PG death. The most rigorous studies of communities in which PG is falling out of use have been conducted by Huffines (1989a,1989b), who looked at the differences between PG as spoken by conservative Anabaptists and other Pennsylvania Germans in Pennsylvania, and Van Ness (1990,1992), who looked at PG in West Virginia (among other places). Van Ness, in documenting the various changes occurring in the PG of West Virginia, concluded that one of the principle factors causing the move away from PG to English was simply isolation from the larger PG community. Similarly, Huffines documented the loss of PG among a group of people in Pennsylvania, and she found that the deciding factor in whether a speaker would maintain PG was whether s/he was a member of a conservative Anabaptist group or not--conservative Anabaptists are maintaining PG, while all others are not. In light of this finding, it is worth noting that all of the speakers in Van Ness's study were non-Anabaptists.
Also worth mentioning in conjunction with these studies is the note by Umble (1994) that a relaxation of OOA religious laws to allow the use of telephones tends to increase the proportion of English to PG spoken by individuals. Although this is based on anecdotal evidence rather than anything quantitative, it is an interesting bit of support for the idea that the more conservative the community, the more likely the members of that community are to have uses for PG rather than English.
3.1.2. Recent studies of linguistic pressure on and maintenance of PG. Several studies have looked at the pressure that PG is under from English in various places in which PG is still actively being used, and at the efforts of PG speakers to maintain the use of PG. Some of these--Clausing 1986, Dow 1988, Enninger 1988, Johnson-Weiner 1989, Huffines 1988, Van Ness 1993,1994, and Thompson 1994--are described fairly quickly in the following paragraphs.
220.127.116.11. Clausing (1986). Clausing's (1986) study does not restrict itself to PG (and in fact does not cover PG among Anabaptist groups much at all), but covers both German and Icelandic in North America. It is however very useful for seeing what other closely related languages have done in reaction to pressure from English, as well as in some cases showing what other groups have done to cope with the loss or maintenance of their non-English language. Such cases are especially important in relation to PG-speaking communities that are losing their PG, as many of the formerly German- and Icelandic-speaking communities Clausing reports on went through very similar sorts of responses to linguistic pressure from English that such PG-speaking appear to be going through now.
18.104.22.168. Dow (1988). Dow's (1988) study is a quantitative review of ways in which English has put new or renewed linguistic pressure on PG in the past few years. Dow presents various proposals for ways in which English is placing pressure on PG in school, social, business, and printing contexts, along with ideas for future studies of these contexts in PG-speaking communities. In the end, Dow is rather pessimistic regarding the future of PG as a viable language.
22.214.171.124. Enninger (1988). Enninger (1988) concludes that the OOA are able to maintain PG in the face of pressure from English because each variety has specific roles in which it is used, and that these roles do not overlap.6 It is worthwhile to compare this situation with the one described by Johnson-Weiner (1989) (described in 126.96.36.199), who studied an OOA community and found that the separation of roles for each language is breaking down and PG is not being completely maintained.
188.8.131.52. Johnson-Weiner (1989). Johnson-Weiner (1989) studied two OOA communities in which the separation of linguistic roles noted by Enninger (1988) had broken down over time; perhaps as a result there is an inverse relationship between proficiency in PG and age. Responses to surveys taken in these communities also support the idea that if a speaker sees no reason to continue using all of the languages in a language contact situation, s/he will tend to discard the language(s) seen as superfluous.
184.108.40.206. Huffines (1988). Huffines (1988) studied conservative Anabaptist PG speakers and discovered that certain convergences to English were in progress, namely the loss of the dative case, use of the verb duh 'to do' as an auxiliary verb,7 and the use of English word order. Huffines concluded was that these convergences were an attempt to maintain PG as a viable language and at the same time cope with the surrounding English-language environment. In this conclusion she stands against Dow (1988), who tended to view any accommodation to English as a sign of the impending death of PG.
220.127.116.11. Van Ness (1993,1994). Van Ness's (1993,1994) studies of changes occurring in PG show changes occurring that might be interpreted as signs of incipient language death through pressure--changes in the lexicon and the pronominal system, apparently under the influence of English. Van Ness, however, points out--as does Huffines (1989) in a study of an area in which PG actually is being discarded--that these changes are not necessarily signs of language death, but are in fact the sorts of changes that go on in healthy languages all the time. Van Ness goes on to show that in the communities she studied PG is not in any imminent danger of being abandoned.8
18.104.22.168. Thompson (1994). Thompson (1994) studied a community in which two dialects of PG have come together, and found that English is occasionally used between PG speakers in order to avoid misunderstandings as a result of dialect differences. This casts doubt on the conventional wisdom that states that all varieties of PG are easily mutually intelligible.9 Despite this use of English as a sort of lingua franca, Thompson found no reason to conclude that PG is falling into disuse in this community, although it bears continued watching.
3.2. Recent studies on the English of the Pennsylvania Germans. The question has at various times arisen of whether PG speakers speak English with a PG-influenced accent, as Frey (1945) asserted, or whether they speak English indiscernibly from non-PG speakers, as Hostetler (1993) has maintained. Certain researchers have looked at this question, sometimes approaching it quantitatively; among the studies conducted are those of Raith (1981), Enninger et al (1984), Huffines (1986), and Huffines (1990). As has been done before in this paper, each of these studies will be presented here separately. Some of them will, however, be presented somewhat more comprehensively than the earlier ones.
3.2.1. Raith (1981). Raith's (1981) study of phonological interference in the English of PG speakers is the one that all of the more recent quantitative studies of the phenomenon start from. The study involved having various PG speakers all read the same English text, noting ways in which the speakers' pronunciations differed from Standard American English pronunciation.10 The results of this test gave an implicational scale of twelve phonological interferences, so that someone who had the first one would have all twelve, someone who had the fifth one would also have the sixth through twelfth but not the first four, etc. In order to give an idea of what interferences were found, the list is reproduced in (1).
|(1)||Standard English||PG-influenced English|
3.2.2. Enninger et al (1984). This paper was a product of the Essen-Delaware Project Team, led by Werner Enninger. The study looked at phonological features of a group of OOA PG speakers in an attempt to verify Raith's (1981) findings; the study also looked somewhat at morpho-syntactical and lexical interference, but the results in those areas were quite preliminary and tentative.
In order to test phonological interference, two tests were used. The first was a reading of a text (not the same one used by Raith in his 1981 study); the results of this were that there was very little PG interference in the speakers' English, and what interference there was occurred sporadically. In addition, although the interferences which occurred were mostly (though not entirely) the sorts that occurred at the higher end13 of Raith's implicational hierarchy, the interferences that occurred did not fit the implicational system Raith laid out.
The second test of phonological interference used a variant of the matched guise test. A tape was made with 9 OOA and 9 non-OOA speakers arranged randomly on one side of the tape, and the same 9 OOA speakers arranged randomly again on the other side of the tape. Listeners (all monolingual English-speaking) were asked to judge whether each speaker was a member of the OOA (and therefore a speaker of PG) or not after having been told that the tape contained samples of both types of speakers. The conclusion drawn was that there was no consistency in rating whether a speaker's English was affected by OOA, but there are two major criticisms of this test and the presentation of the results that would have to be dealt with before this conclusion could be taken at face value.
The first is that all of the results from the second side of the tape (the one that contained all OOA speakers) are suspect; given that listeners were told that the tape contained both OOA and non-OOA speakers, they might have expected that to be true of each side of the tape rather than just the entire tape taken as a whole. The second criticism is that not all of the results were presented; that is, the percentage (total numbers were not given) of the time that all nine OOA speakers were identified as OOA speakers was given, but the percentage of the time that the non-OOA speakers were mistakenly identified as OOA speakers is given for only three of the nine. If all of the percentages were given, one could tell whether there really was a difference in identification--and if the number of listeners were given, it would be possible to tell whether the difference in identification was significant. As it is, such conclusions cannot be drawn from the data presented.
With these criticisms of the phonological portion of the study, and the fact that the other portions of the study were preliminary, not much in the way of hard information can be drawn from it. However, the article offers some useful ideas about ways to test for the existence of PG-influenced English.
3.2.3. Huffines (1986). Huffines did not deal with segmental phenomena in her 1986 study of PG-influenced English; rather, she studied the intonation patterns of speakers of various levels of ability in PG (from fluent PG-speaking to monolingual English-speaking) from an area of Pennsylvania in which PG is passing out of use. She found is that speakers in that area use two different sets of question-intonation patterns, one the standard one for the north midland region and the other a nonstandard one involving such things as falling intonation at the end of a yes-no question, with the nonstandard pattern being used slightly more than half the time.
Upon further investigation, it was found that PG uses the intonation pattern that appears in this area as the nonstandard one; this points to the conclusion that the intonation pattern is a remnant of PG. Interestingly, though, even the monolingual English-speakers from the area use the nonstandard intonation pattern, so the phenomenon is not simply a phenomenon of PG interference.14 Huffines (1986) concludes that the intonation pattern is a marker of ethnicity, and thus is able to survive.15 It would be most useful to test people (both ethnic Pennsylvania Germans and others) from this area to determine whether this actually is an ethnic marker, or whether it is a local variant that is perhaps in some other way the result of the area's historical English-PG contact.
3.2.4. Huffines (1990). This study differs from the others in that it does not look at English as spoken by Pennsylvania Germans, but rather it looks at the languages (mainly English, but also PG) spoken by the Pennsylvania Germans as they are viewed by others (and by the tourist industry). In this study, Huffines (1990) looked at several of the booklets sold in tourist shops in Pennsylvania German areas and points out that they promote long-standing linguistic misconceptions about the Pennsylvania Germans. These misconceptions take various forms, such as the idea that all speakers of PG (and PG-influenced English) are Amish, that PG-influenced English is the result of poor grammar (rather than simply being influenced by nonstandard rules), and that such unconfirmed-by-fieldwork constructions as the infamous "Throw the mama a kiss the train" are common among the Pennsylvania Germans. These and other bits of this supposed PG-influenced English, both performed and written, tend to portray the Pennsylvania Germans as naïve, humorous, and non-threatening.
The conclusion drawn is that this misinformation serves to "validate the prevailing misguided view that English is superior to other languages, that anyone in the United States should and must speak (only?) English, and that minority languages cannot serve the communicative functions of any American community and effectively eliminates vestiges of competitive diversity" (Huffines 1990:124). This may be overreaching a bit, but it is certainly the case that playing on outsiders' (often preëxisting ) misconceptions about the Pennsylvania Germans must serve some sort of culturally significant function. Whether it serves the function of the sort of cultural warfare Huffines describes merits a closer study than the cursory overview she gives it.
4. Conclusions. This paper has attempted to give an overview of the work that has been done in studying PG and the other languages of the Pennsylvania Germans over the past 125 years. In looking at what has been done, two things become apparent--more work on the English of the Pennsylvania Germans is needed, and more quantitative work is needed. In addition, it would be interesting to compare linguistic studies of PG communities with anthropological studies of those communities, because it is apparent (see, for example, sections 22.214.171.124-126.96.36.199) that cultural attitudes have a significant bearing on the use of PG in PG-speaking communities. Also, regrettably, no studies have been done (or at least the results of none have been published) which look at socialization patterns between PG speakers and non-PG speakers (as well as between conservative Anabaptists and others) in relation to retention of PG and PG influence on English.16 Future research taking these things into account will not only help fill gaps in our knowledge of PG-related issues, but will also help resolve issues relating to language contact situations in general and the way that socialization patterns across language lines affect the languages involved in such cases.
1 The other main surviving North American German languages are generally called Mennonite Low German, Texas German, and Wisconsin German, with some disagreement over whether Wisconsin German is actually separate from PG. Mennonite Low, Texas, and Wisconsin German deserve treatment separate from PG, and therefore will not be dealt with in this paper. Reed (1971) has suggested the blanket term "American colonial German" to cover all of these languages, a term which (unfortunately, in my opinion) has not gained widespread acceptance. return
2 This is actually a somewhat inexact usage; technically, one must not be a speaker of PG to be called a Pennsylvania German. The correlation is close enough, however, for present purposes, although as will be seen, some of the studies described in this paper do not hold to such a definition. return
3 Interestingly, the Pennsylvania counties listed were Lehigh, Berks, and Lebanon Counties; Lancaster County, now thought of as the heart of PG, was not included. A likely reason for this is that the three counties listed in the article held a large number of non-Anabaptist PG-speakers, and many dialectological studies of the time tended to ignore the Anabaptist PG-speakers (see, for example, the opening maps in Reed and Seifert 1954). return
4 For example, Struble (1935) offered such items as the switching of [w] and [v] (Wicar of Vakefield 'Vicar of Wakefield') and some non-standard syntactic constructions (He climbed the fence over 'He climbed over the fence'); Page's (1937) study noted such things as some differences in lexical usage (machine 'car') and certain intonational patterns, but nothing at the same level as what Struble reported. return
5 Others challenge Frey's (1945) conclusions in other ways, for example questioning whether PG and Amish High German should be considered separate languages. For a short but thorough overview of the controversy see Rein 1977. return
6 Basically, strict diglossia in the sense that Ferguson (1959) presented it. return
7 One cannot consider the use of duh as an auxiliary in and of itself as an example of convergence to English, as various dialects of European German use tun 'to do' as an auxiliary verb, as well. The article, however, points out that the use of duh in PG is moving closer to English usage. return
8 Although going into much detail on the topic would be beyond the scope of this review, it should be noted that this relates directly to the existing literature on leveling. It would be most useful to look at the leveling (or, perhaps, apparent leveling) occurring in some PG-speaking communities in the light of studies done elsewhere. return
9 I should probably note here that an Old Order Mennonite man (who wishes to remain nameless) with whom I spoke earlier this year about the status of PG also noted that there are significant differences not just between the PG spoken by Amish and Mennonite speakers, but also between communities within the same branch of Anabaptism--and that the differences are great enough to cause occasionally severe problems in communication. It would appear that even if Van Ness (1990) is correct and PG was at one time a homogenous or near-homogenous language, dialect differences and the mutual intelligibility of PG dialects are an important issue that should be looked at more closely in the future. return
10 No information was given on the way Standard American English pronunciation was defined. return
11 is the symbol used in the original table; in more standard usage the symbol would be ow or ou. In any event, the change signified by this line is simply a monophthongization. return
12 Gillian Sankoff (p.c. 1996), in looking at this finding with me, conjectured that it might be the result of the groups with less contact with non-PG-speakers keeping their PG and English more separate than those in other groups. Such a hypothesis would require a great deal of testing. return
13 That is, the items marked with higher numbers in the table reproduced in (1). return
14 This brings up a theoretical question which is well beyond the scope of this paper--at what level does such interference occur? As presented here, the interference is assumed to be internal--a sort of intraindividual interference. However, at some level there must be interindividual interference occurring as well. Where exactly the line lies is an important question for future research. return
15 Interestingly, Zsuzsanna Fagyal (p.c. 1997) has informed me of a study she conducted among the Donauschwaben (ethnic Germans) in Hungary, in which she found that Germanic intonation remains in use by even native Hungarian-speaking members of that group in certain situations. It would likely be useful to compare the two situations and see whether the assertion that intonation can appear as a linguistic vestige of ethnicity may, in fact, be correct. return
16 Teenagers, it would seem, would be particularly good to study, particularly within conservative Anabaptist groups, as at that age they are allowed (some would say encouraged) to experiment a bit in the ways of the world at large, which would allow--if not require--a good deal of contact with non-PG speakers (or at least with those whose PG would likely be somewhat different). return