Zazaki: a West Iranian language
« previous post | next post »
In the midst of our ongoing debates about whether Cantonese, Shanghainese, Taiwanese, and so forth are Sinitic languages or dialects, I continually find evidence that the custom of referring to them only as "dialects" is exceptional when compared with linguistic usage elsewhere (e.g., India, Europe, Africa).
Today I came across an Iranian language that I'd never heard of before, Zazaki, although — without knowing it — I probably met some of its speakers in Sweden, where there are many Zazak refugees. Also called Zaza, Kirmanjki, Kirdki, Dimli, and Dimili, Zazaki is found primarily in eastern Anatolia. It belongs to the northwestern branch of the Iranian group of the Indo-European family.
Zazaki does not have an army, navy, or national flag; it only has around 3 million speakers; it has no place on earth that it can call its own country; many of its speakers live in a far-flung diaspora; until very recently it had no written texts; and so forth — and yet linguists accept it as a separate language. Indeed, Zazaki itself is considered to have at least three dialects: Northern Zazaki, Southern Zazaki, and Central Zazaki, and these are divided into many sub-dialects.
Resources for learning about Zazaki:
Yet Another Web Dictionary
Ethnologue
For detailed linguistic analysis, see Ludwig Paul, "The Position of Zazaki among West Iranian Languages" in the online journal, Iran Chamber. (pdf)
In comparison with the highly refined classification of Zazaki, with its small number of speakers, our understanding of the taxonomy of Sinitic languages, with their hundreds of millions of speakers, is pathetically crude. Shouldn't linguists be paying Sinitic the same sort of serious attention that is devoted to Zazaki?
Among many Language Log posts that have touched upon this vital question of whether the mutually unintelligible varieties of Sinitic should be referred to as dialects or languages are these two recent posts:
"Spoken Hong Kong Cantonese and written Cantonese"
(see especially the second part, "Dialect or Language?")
"English is a Dialect of Germanic; or, The Traitors to Our Common Heritage"
[h.t. Petya Andreeva]
Giacomo Ponzetto said,
September 22, 2013 @ 11:19 am
Scholarly debate may be different, but from a layman's perspective Italy seems similar to China in this respect. Italians customarily refer to all Italian languages other than Italian only as "dialects."
This doesn't imply that they are considered varieties of Italian. I believe they are acknowledged to be Romance languages more or less closely related to Italian, with varying but in some cases quite low mutual intelligibility.
They are called "dialects" to recognize and reinforce their limited geographic spread, cultural reach, and social prestige relative to the national language. They are predominantly oral, to the extent that I doubt any of them has a single accepted orthographic convention, even if several have long and reasonably storied literary traditions.
I think it's also fair to say that for a century and a half Italian national institutions have endeavored to eliminate Italian "dialects" and replace them with standard Italian, and since the advent of mass schooling and national television this endeavor has been rather successful.
Jake Nelson said,
September 22, 2013 @ 11:23 am
As I said in the written Cantonese post's comments, "dialect" can pretty easily be given a clear definition ("An intersection of idiolects allowing communication between multiple people."). "Language", however, seems to have only ad-hoc imprecise descriptions. In order to ask "dialect or language", one would have to answer the question "What is the definition of 'language'?". "Dialect with an army and a navy" is glib, meant more as cynical realism than prescription, and doesn't match how linguists like to group them, but it's at least concise and relatively straightforward.
Personally, I'd favor dropping the term "language" from technical usage altogether, leaving it for its far more common informal/customary use, and use the term "dialect family" or something similar as the grouping. ("The Examplese dialect family contains many living dialects, such as West Examplese, North Examplese, and South Examplese. The "Standard Examplese" variety of North Examplese has the most speakers, being an official language of Examplia.")
Zazaki looks interesting. I'm fond of Iranian languages- just enough core Indo-European elements to make them familiar, but still quite different. Not at all fond of grammatical gender, though. Second-biggest stumbling block for a language in my opinion, after phonemic tone.
Jake Nelson said,
September 22, 2013 @ 11:40 am
@Giacomo Ponzetto: Yes! I meant to mention Italy in the previous thread, and completely forgot. I agree that it is very much like the Chinese situation in many ways. Some of the Alpine dialects sometimes referred to as "dialects of Italian" are actually Germanic, but there's that political thing again…
Victor Mair said,
September 22, 2013 @ 11:48 am
Germanic "dialects of Italian" shows how ridiculous things can get when politics are allowed to trump linguistics.
Neil Kubler said,
September 22, 2013 @ 12:47 pm
I always tell my students: a dialect is a speech form close enough to your own speech that you can understand MOST (let's say at least 70%) of it, but different enough from what you speak (in pronunciation, vocabulary, and/or grammar) that you can tell the person speaking it comes from a different region than you do. By this standard, Tianjin hua and Beijing hua are dialects of Mandarin, and Taipei Southern Min and Penghu Southern Min are dialects of Southern Min. On the other hand, Mandarin, Cantonese, Hakka, Shanghainese, Xiang, and Southern Min are different languages, since they are mutually unintelligible. Moreover, the common statement by Chinese and foreigners alike that "different Chinese 'dialects' are pronounced differently but are all written the same" is not correct, since the fact is that Cantonese and Taiwanese have separate writing systems (some using characters and some using romanization) which would not be more than about 50% intelligible to a reader from Beijing. The truth is that first graders in Guangzhou and Hong Kong and Gaoxiong and Xiamen and Suzhou speak one language at home and learn another (written Mandarin, also called Baihuawen or Guowen) in school. (It gets even more complicated, since written Mandarin can be pronounced in Mandarin, as it is in Guangzhou and Gaoxiong, or it can be pronounced in the local language, as it is in Hong Kong.) Now, words mean what you and your interlocutors agree they mean. If to follow Chinese (cultural and political) custom, we want to speak loosely of "the Chinese dialects," I suppose that's OK, so long as in the back of our minds we are clear that they are really different languages.
Steffen Larsen said,
September 22, 2013 @ 1:01 pm
@Jake Nelson: Can you elaborate what you mean by grammatical gender being the "[s]econd biggest stumbling block for a language"?
Many languages distinguish between grammatical genders without suffering from loss of speakers.
Wentao said,
September 22, 2013 @ 1:12 pm
Is it possible to compare the similarity between two languages quantitatively? If so, I wonder how different the Sinitic languages are, compared to the internal variety within Romance and Germanic languages? Which language pair in Europe is comparable in terms of dissimilarity with Mandarin and Cantonese: High German/Low German, Spanish/Portuguese, or English/French?
cameron said,
September 22, 2013 @ 2:18 pm
Zazaki is an interesting case because its speakers are considered, and consider themselves, Kurds. If Kurdistan were a political entity Zazaki might be considered a dialect of Kurdish. Languages similar to Zazaki spoken in Iran are considered dialects of Persian. But most of the speakers of the Caspian languages can code switch into standard Persian, and that makes them seem less distinct.
JR said,
September 22, 2013 @ 3:20 pm
@Neil: But that definition would mean Spanish and Portuguese are not separate languages.
Also, I thought everyone speaks a dialect, even though my Cleveland dialect differs so slightly from the Standard, that no one in California noticed I was from a different region–they didn't notice I had no cot/caught merger, for instance.
Iamaom said,
September 22, 2013 @ 6:08 pm
@Neil Kubler
To add to what JR said.
That definition of dialect is completely subjective on the person's experiences though. As an American whose never been to Britain I'd be forced to call british street slang a different language than mine, while an American whose been living there for some time wouldn't.
Victor Mair said,
September 22, 2013 @ 10:16 pm
From Daniel Mellis:
It seems to me that people often refer to African languages as dialects. Such as the recent article in the New Yorker about Ghanaian-British architect David Adjaye's, which called his native language Twi, a dialect.
Lazar said,
September 22, 2013 @ 11:48 pm
@Victor Mair: But Twi is considered by many learned sources to be a dialect of Akan. A better example would have been if someone had referred to something that's more unambiguously a distinct language, like Zulu, as a dialect.
Stephan Stiller said,
September 23, 2013 @ 3:09 am
@ Giacomo Ponzetto
You wrote:
According to Wikipedia, Piedmontese, Ligurian, Sardinian, Friulian, and Ladin have standardized written forms. But I don't know how widely used these orthographies are.
Giacomo Ponzetto said,
September 23, 2013 @ 4:30 am
@ Stephan Stiller
I was trying to say that Italian dialects have multiple standardized written forms, and even when one is dominant it doesn't quite manage to achieve universality.
Perhaps in some cases the standard orthography is almost universal. But I believe there remains an awareness that you're using such and such a spelling convention, often identified with its particular authors, and that there have been (and there are) alternatives.
On the contrary, Italian has a single orthography that has been stable for long enough that even educated speakers are more likely to perceive it as an intrinsic characteristic of the language than a convention established by certain authors at a certain time.
Admittedly I'm not a linguist, so all this should be taken with the due pinch of salt until an Italian dialectologist can pitch in.
Victor Mair said,
September 23, 2013 @ 5:36 am
It would appear that a number of commenters to this post and previous posts having to do with taxonomy are confused about the difference between "language" and "dialect". In linguistic classification, "language" lies above "dialect", the order of units being thus: family, group, branch, LANGUAGE, dialect, sub-dialect…, idiolect.
The names and order of units above "LANGUAGE" may vary according to different authorities, but two things are clear: "LANGUAGE" lies at the center of this hierarchy, and "dialect" lies below "LANGUAGE". Thus it is odd when attempts are made to make "dialect" or "idiolect" the primary unit of classification, or, worse yet, to dispense with "LANGUAGE" altogether.
For those who are unfamiliar with linguistic classification and the nature of "LANGUAGE" and "dialect", here are some introductory resources:
http://www.indiana.edu/~hlw/Introduction/dialects.html
http://www.nsf.gov/news/special_reports/linguistics/dialects.jsp
http://encyclopedia2.thefreedictionary.com/Language+and+dialect
And last, but not least, I strongly recommend the following ms. by John R. Rickford:
"How Linguists Approach the Study of Language and Dialect"
http://www.stanford.edu/~rickford/papers/l73_reading_1.doc
Rodger C said,
September 23, 2013 @ 8:09 am
About 40 years ago I listened to a conversation between a young academic who referred to "Native American dialects" and an older one who kept trying to correct him to "Indian languages." It was a case of cross-purpose mutual political correction.
Victor Mair said,
September 23, 2013 @ 8:16 am
@Rodger C
Well, they could have compromised and said "Native American languages".
Doug Henning Jr said,
September 23, 2013 @ 9:36 am
I first heard of this language when browsing the MIT OpenCourseWare site. A linguistics class called Grammar of a Less Familiar Language available there uses Zazaki as a model language for various grammatical features. Haven't delved very far, so I can't say with certainty which features they highlight.
Jake Nelson said,
September 23, 2013 @ 9:52 am
@Steffen Larsen: I was thinking more in terms of personal experience than a general rule, but it's essentially the issue of not being able to make a sentence work correctly at all in the beginning that really hurts for learners. A language where you can at least build simple sentences and slowly work your way up to complexity is much less frustrating to learn than one where you're making mistakes or draw a blank on how to proceed.
A stumble doesn't necessarily mean a stop, of course, and I don't mean to say features like grammatical gender are prohibitive, but they are… obstructive to progress.
Victor Mair said,
September 23, 2013 @ 10:08 am
Two important items concerning Sinitic languages and topolects:
The first is an online dictionary of several Chinese topolects and languages. You can put a character or more on the search bar, then choose the language or topolect for which you want to know the pronunciation.
http://cn.voicedic.com/
The second is a report about the decision to establish the United Chinese language system platform. It is aimed at preserving various topolects and languages in Chinese by means of information technology.
http://www.csstoday.net/xueshuzixun/jishizixun/84824.html
I would appreciate feedback on both of these items, particularly concerning the quality of the pronunciations (both transcreiptions and recordings) provided by the first.
Giacomo Ponzetto said,
September 23, 2013 @ 10:08 am
@ Victor Mair
As you explained, the linguistic facts are not controversial. In a linguistic taxonomy, Cantonese, Shanghainese, etc. are languages within the Sinitic family and not dialects of Mandarin.
But is this what the distinction between "Chinese dialects" and "Sinitic languages" is really about? Do people who say "Chinese dialects" mistakenly think that means "dialects of Mandarin?"
In common speech, as opposed to technical linguistic terminology, the distinction between language and dialect may reflect a different type of hierarchy. The OED gives the following definition of dialect:
The last sentence implies there's no logical inconsistency in distinguishing between, on the one hand, standard Mandarin as the national, literary, and official language; and on the other hand all the subordinate provincial Sinitic languages.
So it is possible that people who say "Chinese dialects" don't mean mistakenly "dialects of Mandarin," but rather "inferior Sinitic languages," which may be odious but is neither ridiculous nor confused.
In Italy, at least, everyone says "Italian dialects" and doesn't mean "dialects of Italian" but "inferior languages of Italy," where the inferiority is geographic and socio-cultural. None of the online Italian dictionaries I'm familiar with even provides the definition of dialect that is correct in linguistic classification (Gabrielli (Hoepli), Garzanti, Sabatini-Coletti, Treccani).
Rodger C said,
September 23, 2013 @ 10:12 am
@Victor Mair: I think neither of them ever grasped what the other one was getting at. It was a common situation in those days right after the great culture shift.
Jake Nelson said,
September 23, 2013 @ 10:15 am
And I've said it before, but the language/dialect distinction issue constantly reminds me of the very nearly identical Species Problem in biology.
"Mutual intelligibility" vs. "Reproductive compatibility", dialect continua vs. parapatric speciation (ring species, et al.), prior hierarchal classifications being redefined in terms of cladistics (but still having the "which level to place a group at" issue), horizontal transfer and hybridization (which turn out to be far more common and important than previously thought) presenting a continuous problem for traditional top-down single-ancestor tree models…
Jake Nelson said,
September 23, 2013 @ 10:39 am
@Victor Mair: That site looks quite interesting… I see they have a separate subsite for Japanese kanji readings. Would be nice (from my perspective) if they were in the same place, honestly… A project I'm working on has led me to do a lot of comparing Japanese and Korean kanji/hanja readings versus each other and Mandarin. Something like this with Japanese and Korean readings in the mix right there would be a big help with that, probably. Sites I'm currently dealing with are a bit time-consuming.
Recordings are a little inconsistent, but honestly better than I'm used to in that regard. Overall, pretty nice.
Victor Mair said,
September 23, 2013 @ 11:19 am
Thanks for the assessment of that site, Jake, and I do appreciate what you're saying about the refractory nature of the language / dialect continuum. Still, if we're going to make any sense of how languages (and species) are related, we can't just throw up our hands and say that all of their manifestations are at the same level — e.g., that Camellia sinensis = Camellia, that Cantonese = Hanyu / Sinitic, and so forth.
Mark F. said,
September 23, 2013 @ 1:10 pm
@Victor Mair – I don't think Jake meant at all to suggest the problem was one at which to throw up one's hands. Like you said, biologists don't give up on the species problem, so making the analogy to that case suggests that linguists shouldn't give up on the language taxonomy problem either.
The biologists just have it easier because wolves don't get offended if you decide that dogs are actually a species of their own and not just a subspecies. (Kind of a made-up example, I know.)
Incidentally, it seems like it's politically easier to insist that languages (such as Cantonese) really are distinct languages, than to insist that two highly-mutually-intelligible speech forms of neighboring countries really ought to be considered dialects of the same common language. So that's kind of an interesting asymmetry in the political challenge of trying to be consistent in distinguishing languages from dialects. This also doesn't dispute your main point at all.
Steffen Larsen said,
September 23, 2013 @ 1:28 pm
@Jake Nelson: I can see your point about grammatical gender being an obstructive to progress, especially if the learner's native language lacks this feature. From my experience there is always certain features in a language that will be obstructing progress, but in my opinion that only makes it so much satisfying when you progress past the obstruction.
Norwegian (my native language) has grammatical gender, those being masculine, feminine and neutral gender. That might cause trouble for some learners, but I think getting to grips with norwegian orthography is a much larger obstruction than grammatical gender given the recent changes made. Example: "To wish" can now be spelled in four different ways if I remember correctly.
The number of dialects is another obstacle for foreigners with a desire to learn norwegian, since there is no standardized spoken variety which has gained a significant number of speakers (we are proud of our dialect/native variety/native language).
Steffen Larsen said,
September 23, 2013 @ 1:29 pm
*grammatical gender being obstructive to progress,
instead of: being an obstructive to progress.
julie lee said,
September 23, 2013 @ 1:37 pm
When I was in Hong Kong and Taiwan the years immediately after Mao took over China Mainland in 1949, there were a lot of European Jesuits (French, Belgium, Italian, etc.) at Catholic centers who had converged on Hong Kong and Taiwan after fleeing far-flung parts of China. Most of them didn't know Standard Mandarin but only spoke the topolect (of Shandong, Shaansi, etc.) of the province where they had been stationed for many years. (I also met an Italian Jesuit who said his native tongue was not Italian but Ladin.) Because many of the Chinese topolects these Jesuits spoke were mutually unintelligible, they usually communicated with one another in English. I was in stitches when I heard one of these Europeans give a sermon to the Chinese audience (mostly Standard Mandarin speakers) in a Shansi or Shaansi topolect. (I imagine it would be like hearing a Chinese priest give a sermon in fluent cockney to a cultivated English audience in London.
Victor Mair said,
September 23, 2013 @ 4:45 pm
@Giacomo Ponzetto
Nobody says "dialects of Hindi" for the various Indic languages of India. If they did, people would think that they are loony. I maintain that it is just as loony to refer to Cantonese, Taiwanese, Shanghainese, etc. as "dialects of Mandarin".
J. W. Brewer said,
September 23, 2013 @ 5:10 pm
No, but people do say "dialects of Hindi" for the many not-always-mutually-intelligible language varieties whose speakers are lumped together for political purposes by the Indian government as "Hindi," even though some of these topolects might be better thought of as independent languages (e.g. Rajasthani or Bihari) yet are not given the sort of political/social recognition that e.g. Bengali gets. A narrower definition of Hindi would reduce the percentage of the population claimed to speak it, which might create political awkwardnesses. Think of Hindi in this sense as loosely parallel to the variety of not-always-mutually-intelligible language varieties lumped together in the PRC as "Mandarin," with the government-approved Standard Hindi playing the role of MSM. Admitting that "Bengali" is as separate from Hindi as Cantonese is from Mandarin is obviously a good first step. (But what's "Bengali"? Apparently many/most immigrants from Bangladesh in the UK speak Sylheti rather than the standard literary-norm version of Bengali, and there is of course debate as to whether Sylheti is a "language" or merely a low-status "dialect" of Bengali.)
Victor Mair said,
September 23, 2013 @ 8:49 pm
When people say "dialects of Hindi", we know that they're referring to varieties of Hindi, not to Bengali, Gujarati, Marathi, Oriya / Odiya, and so forth. When sensible people say "dialects of Mandarin", they should be referring to varieties of Mandarin, not to Cantonese, Taiwanese, Shanghainese, Amoyese, Quanzhouese, etc.
You may quibble about the borders of Bengali, but, as I've pointed out before, it has a long history and a glorious literary tradition.
http://en.wikipedia.org/wiki/Bengali_language
http://en.wikipedia.org/wiki/Bengali_literature
If you don't know what Bengali is, read some Tagore. Or, better yet, listen to some, as in the embedded video here:
http://fortnightlyreview.co.uk/2013/04/rabindranath-tagore/
Jake Nelson said,
September 24, 2013 @ 1:44 am
@Mark F.: yes, indeed. Classifying people and their cultures is a lot more perilous (and ethically fraught) than classifying things that don't have opinions about the process (as far as we know).
@Steffen Larsen: Indeed, I tried to learn Norwegian when I was younger (Most of my ancestors were Norwegian, and a lot of old Norwegian books have been passed down), and was more stymied by inconsistency than the actual grammar.
@Victor Mair: While your point about dialects of Hindi makes sense, whenever someone says "Nobody says "[x]"" , I immediately think "Someone, somewhere certainly does.". I'd agree it's probably incorrect to do so, but that's a different matter. Just saying…
Victor Mair said,
September 24, 2013 @ 5:31 am
@Jake Nelson
"Nobody [who thinks half straight] says '[x]'".
Anybody who says that "Hindi dialects" refers to Bengali, Gujarati, Marathi, Oriya / Odiya, and so forth is not thinking half straight.
Rakau said,
September 24, 2013 @ 11:17 am
Polynesian languages don't seem to suffer from the language/dialect problem. New Zealand Maori and Cook Island Maori are very similar and could easily be considered dialects, but aren't. As far as mutual intelligibility is concerned polynesian languages shed interesting light on our origins and pre-european migrations. A fluent speaker of New Zealand Maori will understand more of Hawai'an, Tahitian Maoli and Easter Island Maori (not sure here what the name is) than all the languages of the islands in between e.g. Samoan, Tongan and Niuean. The languages at the extremities of the old polynesian triangle are mutually intelligible while those in the middle are not.
Victor Mair said,
September 24, 2013 @ 12:09 pm
@Rakau
"The languages at the extremities of the old polynesian triangle are mutually intelligible while those in the middle are not."
It's very interesting that the law of conservative peripheries and innovative centers seems to hold even in the vast, sparsely populated Pacific.
Greg Morrow said,
September 24, 2013 @ 12:49 pm
My native language is non-gendered (English), and in my limited experience (French and Spanish), gender was not a significant impediment. In most cases in both languages, it's predictable, and where it's not, it's not a substantially greater load on the lexicon to load the gender in with the rest of the entry (sound and syntax and meaning and spelling). But I have a pretty good lexicon.
Admittedly, my second language lexicon is full of cheats — to this day, my lexicon entry for the French verb cognate to "obey" is "obéir à" — we were taught to remember it that way so as to remember that the verb takes an à prepositional phrase instead of a direct object.
Simon P said,
September 25, 2013 @ 12:36 am
"I would appreciate feedback on both of these items, particularly concerning the quality of the pronunciations (both transcreiptions and recordings) provided by the first."
I tried the link with a few characters and Cantonese. My observations:
1: Doesn't work with traditional characters (minor flaw)
2: It works with some but not all Canto-specific characters. 畀 doesn't work, but 佢 does. 唔 is given as "ng4" which I believe is a hypercorrection. Everyone pronounces it as "m4".
3: It's not completely accurate. I didn't try many characters, but I noticed 肯 is gives as "hoi2" which is incorrect. Should be "hang2".
Simon P said,
September 25, 2013 @ 2:26 am
Oops, looking again, it looks like the first reading "hang2" was obstructed by that floating QR code. The "hoi2" reading is probably some obscure ancient variant or something. So it looks pretty good.
Janet Williams said,
September 25, 2013 @ 2:47 am
@ Victor Mair:
I tested Hokkien on 汉语方言发音字典: Voicedick. I wonder how representative the sounds are. In Malaysia and Singapore, Money (钱) is more commonly pronounced as 'lui' than [zi2], due the the Malay influence. In the Malay language, Duit is money, and duit was also a copper Dutch coin. It is likely that the Dutch rule brought the word 'duit' into Indonesian, and the pronunciation is later naturally absorbed into the Hokkien spoken in Malaysia and Singapore. I only pronounced [zi2] for 'money' when in Taiwan.
Stephan Stiller said,
September 25, 2013 @ 6:14 am
@ Simon P
I think for your browser or phone, the floating QR code obstructed the entire first column. I get pronunciations for all of 畀佢唔肯: bei3; heoi5, keoi5; m4, ng4; hang2, hoi2. I would say a good database will display: bei2; heoi5, keoi5; m4; hang2. Though you can argue about whether to include 畀/bei2.
That the database contains pronunciations that aren't used is one problem; that it lacks some vernacular pronunciations is another. I won't give examples here.
A bigger problem is that vernacular morphosyllables without a commonly agreed-upon character might not appear in such a database. The database also doesn't try to disambiguate by word.
Giacomo Ponzetto said,
September 25, 2013 @ 7:29 pm
@ Victor Mair
You are absolutely right: calling Cantonese a dialect of Mandarin is crazy — like calling Walser a dialect of Italian, or Bengali a dialect of Hindi.
I wonder if there was ever a custom of calling Bengali, Gujarati, and maybe even Hindi, "Indian dialects," to imply that only English and Sanskrit were proper high-culture languages. Google Books returns a few colonialist gems such as:
(The Asiatic Journal XXV, May 1828)
But these seem rather few and rather old. Maybe even the Raj was more respectful of Indian language variety than that.
Twana Hamid said,
September 29, 2013 @ 4:55 pm
Linguistic criteria are not the only factor in determining two groups of speakers as two dialects or two languages. It is assumed that speakers of two dialects (of the same language) are mutually understandable. However, in the case of the dialect of Zazaki, it is not mutually understandable by other Kurdish dialects.
Sorani, Kurmanji and Zazaki are classified as Kurdish dialects as the speakers of these dialects regard themselves as Kurd, not due to understanding each other.