Talk:Internal reconstruction

Revision of the article
I'm undertaking a complete revision of this article. As written, it can hardly be understood by anyone who doesn't already know what the term means, which is a demerit in a reference work. Alsihler 22:33, 28 January 2007 (UTC)


 * Alsihler: I'm glad to see some examples added to the page. Thanks for doing that. However, I feel like your rewrite had the overall effect of making the article much more difficult for a novice to understand, and as such I plan to make some changes. One change will be to delete this segment of the first paragraph:
 * The comparative method's goal is to see if such patterns of similarity and difference can be explained most economically by the theory that two (or more) languages were formerly a single language, as indicated by (a) plausible surmises as to the structural details of the basic elements of that proto-language, as it is called, and (b) a collection of regular or otherwise defensible developments through time that account for the patterns of differences and similarities as seen in the attested languages.
 * As this is an article about internal reconstruction rather than about the comparative method, it seems inappropriate to give so much attention to the comparative method. It seems that your purpose was to contrast the comparative method and internal reconstruction, so I'm going to restore the sentence from the previous version that did just that:
 * Whereas the comparative method compares variations between languages — such as in sets of cognates — under the assumption that they descend from a single proto-language form, internal reconstruction compares variant forms within a single language under the assumption that they descend from a single, regular form. For example, these could take the form of allomorphs of the same morpheme.
 * Overall, I intend to eliminate the "textbook" / tone that speaks with unjustified authority on the topic, and makes irrelevant subjective claims, all without sufficient supporting citations. This could be regarded as original research which is, of course, against WP policy. Also, the goal is not so much to create a handbook or instruction manual on how to become a gifted practitioner of internal reconstruction, but rather to write an encyclopedia article. Examples:
 * In the case of languages whose histories are well understood, either via the comparative method or historical attestation of significant time-depth, internal reconstruction is little more than an entertaining parlor-game, at best a kind of test to see if the data and the reasoning applied to them actually "work";
 * This is an interesting point, and an insightful one,...
 * This is in part from necessity: the data themselves are basically different in the two types of historical analysis. The whole of a language is the comparative method's arena; internal reconstruction is practically limited to those components of a single language which obligingly exhibit alternation,
 * To a historian's eye, it is obvious what is going on, here.
 * Anyway, what you've written is very impressive and shows that you are clearly well-versed in historical linguistics methodology. The problem is that it's out of place both in content and in style. Perhaps, rather than having four in-depth examples with painstaking analyses we could select one example to give a more general illustration. If there is still a desire to get into more detail, we could do so under sub-headings that quickly identify some common issues in internal reconstruction. For example, the discussion of neutralizing environments.
 * By the way, what is the source of your examples and analysis? If these are your own work, they should probably be removed in favor of using examples from an established source, such as Campbell.
 * I'm going to delete all of the "Note" sections, restore the original introductory paragraphs, and do a bit of reorganization for now, but there will still be much to do. Let me know if you have any thoughts. Thanks! --RockRockOn 23:29, 5 July 2007 (UTC)

Well, at least we both seem to agree that 45-odd years of studying, teaching, and publishing on historical linguistic matters has resulted in some level of competence.

I have to confess that all the criticisms of my handiwork baffle me. First, the dividing line between a "'textbook' tone that speaks with unjustified (eh?) authority" and an "encyclopedia article" is too faint for these old eyes to see clearly.

I'm afraid that the examples you give of what you see as faults in presentation/discussion/tone, leave me quite at sea over what the problem is. It has always been my preference to try to engage the reader in a sort of conversation, while at the same time maintaining appropriately rigorous technical accuracy. A colleague (who, as it happens, sits in the oldest chair of Indo-European Studies in all of Europe) flattered me by remarking, after reading a recent monograph of mine on Edgerton's Law, that he is always impressed by my ability "to write interestingly even about dry subjects". So might I suggest that before you rewrite anything, you get a second (or third & fourth) opinion? Incidentally, so far from anything herein being "original research", it is in fact all perfectly standard handbook stuff, well known to anyone with professional experience in diachronic linguistics, leavened by lessons from experience both in and out of the classroom.

Finally, I'm quite at sea over the query about the "source" of examples, etc. Examples are just that, examples. There is no canon of examples in linguistics, except for a few chestnuts (like Armenian erku < PIE *duwo to show dramatically remote phonological correspondences). Otherwise, different authorities choose examples that appeal to them for whatever reason, some of which are encountered elsewhere, some not. There is no difference between the two types. I don't recollect, for example, ever seeing English pew cited as a cognate of English foot (it's a loan from a French reflex of Latin podium), but if one were to do so, it would not be "original research": all the connecting links etc. are to be found in published sources: handbooks, etymological dictionaries, and so on. Nor is a published example by definition beyond reproach. In his recent book Pre-Indo-European (and I sincerely hope his last), W.P. Lehmann undertakes to explain internal reconstruction for the benefit of his non-linguistic readers. The page is utterly incomprehensible even if you know what he's trying to say, which is likely to be the reaction of an uninformed reader, too; and the specific example he presents at the end is both (a) not an example of internal reconstruction in the first place and (b) every bit as incomprehensible as the passage preceding it. Differently, Terry Crowley's introduction to historical linguistics is a useful and well-written book, and delightfully refreshing because most of his examples are from Austronesian languages. But: almost every venture into (Indo-)European languages is a botch -- incorrectly cited forms, faulty analysis, complications that he's unaware of, and so on. Put differently, citing Crowley's "examples" simply because they've appeared in a printed book would be in the last degree unwise.

I'm sorry to be so grumpy, but you probably can imagine why. To repeat what I've already said, you really ought to shop around a little to see if others share your opinions about the style and manner of my article. Not long ago, a former student of mine recommended my book Language History to a friend of his, who reproached him later for making him lose a whole night's sleep because he couldn't make himself stop reading. I don't think it's overweening pride to suspect that fascination with the subject was not the whole explanation (if it was, he should steer clear of Hans Hock's introduction to historical linguistics, which would likely prove fatal). Alsihler (talk) 00:59, 2 February 2008 (UTC)

That said, I'm deleting the following:

===Indo-European=== Some comparative Indo-European linguists, such as Winfred Lehmann, use the term "Pre-Indo-European" to refer to an earlier stage of the Proto-Indo-European language, reached through the method of internal reconstruction.

By using the method of internal reconstruction an earlier stage can be determined. Some linguists, such as Lehmann (2002), propose they can reconstruct aspects of a Pre-Indo-European language. In his book with this title Lehmann has reconstructed the lexicon, syntax and morphology of such a language. There is also a discussion of its phonology and the culture of its speakers. It is suggested that the people were a pre-ceramic Neolithic society.

Lehmann's theories on these points do not merit perpetuation; and the passage is far too vague and allusive to be of any use to anyone in any case; and is poorly written besides. (I do not remember writing it, and Lehmann actually makes the most minimal use of Internal Reconstruction in framing his theories.) Anyone who'd like a more detailed discussion of the shortcomings of a very bad book, please see my review of it in Diachronica which has the merit of being a good deal more accessible than a monograph published by JIES. Alsihler (talk) 22:35, 2 February 2008 (UTC)


 * Note to Dbachmann: not clear what the reference to n ~ ṇ is; apparently someone has deleted the passage. (And in general the whole article makes less sense than I believe it did, if I may say so, when it was first written.) If the reference is (was) to Sanskrit phonology, I beg to differ: /n/ and /ṇ/ were most definitely different phonemes, if not to begin with, of course. Already in the Rigveda there forms with "unmotivated" ṇ, and n where cerebralization would be expected (e.g. guṇa- on the one hand and something like varṣánirṇijaḥ in RV 3.26.5c, where the -n- is unexpected. To be sure, across compound boundaries cerebralization is unpredictable, tending not to occur but in say the inflection of vṛtrahan- it is quite regular. (It can hardly be argued with a straight face that the compound status of the word was opaque to Sanskrit speakers.) The preverb sequence pra-ni- sometimes cerebralizes, sometimes not. The same goes for enclitics, like bhara naḥ "carry us" next to bhara ṇaḥ and so on.Alsihler (talk) 18:40, 8 February 2008 (UTC)

Style, tone and complexity
This is less a Wikipedia article than it is an original-work treatise by a very knowledgeable expert who knows how to write essays and books for consumption by other budding experts; but quite clearly not for a merely interested reader who might happen by. The style and tone in particular are out of place, featuring narrative rhetoric and colloquialisms such as "Now, ...", and suffering at times from Wall-of-Text syndrome. GavinZac (talk) 14:58, 27 October 2010 (UTC)


 * I agree. The text is informative, but it needs its tone reworked. Ke6jjj (talk) 04:09, 6 December 2011 (UTC)


 * I suppose by now you've succeeded in dumbing this down? Thanks loads!  Do you know how much money a person can spend on decent linguistics literature??  And Sihler is a master in his field!  What makes you think everyone who turns here should be "merely interested"?  -- Your complaints seem to be saying, "This is too hard for me/us (or my/our friends)."  Have you also gone and watered down the math, chemistry and physics articles so that they don't strain you?  And if not, why not?  Why is it okay to draw a mustache on this Mona Lisa but not on them?  Your attitude is very much unappreciated.  :~[   IfYouDoIfYouDon&#39;t (talk) 08:36, 27 February 2013 (UTC)

Why style, tone, and complexity matter
(In response to IfYouDoIfYouDon't)

The information in this article is important. It is written in beautiful, flowing prose by an obvious expert in the field. It is understandable to those relevant in the art. It just doesn't conform to the norms of a good Wikipedia article.

It needn't be deleted nor dumbed down; it simply needs to be restructured.

Wikipedia is meant to be a trustworthy starting place on just about any subject. To be trustworthy it must be verifiable—meaning that any user should be able to cross-reference the statements made in the article against another trustworthy source. To be trustworthy it cannot be the sole source of a statement, it must reference other, published, material. To be a starting place no single article can delve too deeply into any subject without risking being unapproachable. These are not my requirements, nor are they GavinZac's requirements; they are Wikipedia's core principles (see WP:VERIFY, WP:ORIGINAL).

This masterful article, as it currently stands, violates these principles. This is a harsh statement. However, is not a disagreement with statements such as "Sihler is a master in his field!". It is not a call for "dumbing this down" so that it can only be read by someone who is "merely interested" in the subject, nor is it an exhortation to "draw a mustache on 'this' Mona Lisa". On the contrary! This article simply needs to be restructured.

Here's what needs to happen:


 * 1) The article's tone needs to be re-worked.
 * 2) It needs inline citations.
 * 3) It needs to be broken into more sections.
 * 4) It may need to be broken into more articles.
 * 5) The lead in needs to be shortened and summarized.
 * 6) The text might be better published as a Wikibook. (See WP:TEXTBOOK, "Wikipedia is not a textbook")

Re-working the tone
The article's tone (who its writer pretends to be, whom it purports to talk to, how it introduces factual data) needs to change.

It does not take a neutral point of view.

When undertaking a comparative study ... it is worthwhile... Lyle Campbell...raises an interesting caution but a basic principle of linguistic analysis is that one cannot analyze data they do not have (and should not try to)This is an interesting point, and an insightful one...

It directly addresses the reader, changes accumulate in the structure of a (living) language, and for this reason we always try... We have two choices if we stick to the data:we can write an unambiguous rule for the infinitive forms:

and even poses pedagogical questions. Can we make any generalizations about the membership of verbs in Types I and II?

All of these examples make great copy for a textbook! I would surely read more of Sihler's texts, if I had them. The examples aren't, however, acceptable in a reference work. Wikipedia seeks to make the writer nearly invisible and pretends that the reader doesn't even have a body. A day may come when Wikipedia is engraved on the surface of an asteroid and read by an intelligent, alien race of machines who seek to understand our long-lost civilization; words like "we" and "you" make a lot of assumptions about them.

Statements that something "is interesting", or that something "should not be done" are not neutral in tone. They advocate a position, however uncontested it may be, and assign subjective values that can't be quantified. They don't make good encyclopedic facts. What would make a good statement? Perhaps something like,

Linguist Andrew Sihler disagrees with the practice of...

and so on.

Adding in-line citations
There is one in-line citation provided in this entire article and I must admit, it is a miserable link to a Powerpoint presentation of a class outline of linguistics course at UNC that I added mere hours ago. This article contains hundreds of blanket statements which deserve citations. It needs citations not just to inject more authority into the discussion, but also to help interested readers explore the ideas behind the statements in more depth. They might introduce the reader to other books and websites dedicated to the subject. Without further clarification, we have to assume that all of the text in this article is just Sihler's original research.

Breaking into more sections, sub-sections
As GavinZac noted, this article suffers from "Wall-of-Text syndrome". It engages the reader for paragraphs and paragraphs (nine in one case) without taking a break to produce a heading, or any other navigation aid, that allows an interested reader to skip over sections they already are familiar with or allows a hurried (but knowledgeable) reader to jump into a section containing information they really need. In short, it's just hard to navigate.

Wall-of-Text is curable. The cure comes from creating sections and sub-sections. Introduce a topic in a section, then show examples, counter-examples, and debates within. Move on to the next topic with a new section.

Breaking into more articles
Sometimes a concept is so complex that it warrants breaking the article into several articles. This subject could be one of them. Perhaps "Latin" and "English" under the "Examples" section could become new articles: "Internal reconstruction of English" and "Internal reconstruction of Latin".

Shortening the lead in
The article's lead in needs to be shortened and summarized. The sentences that read "For example, ..." need to be moved elsewhere. An introduction is meant to introduce, not to tell a life story. Examples are great...in later sections.

Publishing as a Wikibook
Finally, if all of these fixes are just too hard to bear, consider moving the article, as is, into a Wikibook. Wikibooks are maintained by the Wikimedia foundation, have tremendous prestige, and are free (as in beer) for all to read for countless generations. This article is already in textbook style and would make a great start as a Wikibook.

Ke6jjj (talk) 01:31, 9 January 2014 (UTC)

Spanish Spelling
Why has someone messed up the spelling in the Spanish example? If you want a phonemic spelling, use the IPA. There's no need to go messing around with orthography.101.166.150.53 (talk) 14:07, 22 December 2012 (UTC)

Samoan examples
As of revision 577155539, the article still references examples in Samoan, which were removed in 2010. I propose that the references go. LLarson (talk) 15:54, 8 December 2013 (UTC)
 * Either that or some expert(s) rework it entirely. Citations are needed anyway. 151.177.58.208 (talk) 12:20, 18 January 2022 (UTC)

-t and -ted verbs
meet and bleed actually are denominal verbs, though meet is much more common than moot in modern English, and read is a strong verb, so it shouldn't be in this list. In German, senden is in both classes. phma (talk) 23:03, 9 August 2019 (UTC)