Definition from Wiktionary, the free dictionary
Jump to navigation Jump to search

Vocative Singular[edit]

The source of the information regarding the vocative singular in Classical Greek would be my Greek TA from my elementary Attic course. It would be better, however, if someone could find an additional source. My apologies for not having one. Medellia 03:59, 13 October 2006 (UTC)

Contradictory etimology[edit]

"Despite the superficial similarity, the word is not related to Latin deus" -- this is nonsense. Of course those words cognate, hence the phrase above "Cognate with Phyrigian δεως (deōs, “to the gods”)". It is clear that Greek and Latin words are not borrowed from each other language, but those are related from common Ingo-European roots. 11:29, 16 May 2012 (UTC)

Do you have a source to back up this claim? The IE roots look fairly different to me (*deiw- for deus and *dʰ(e)h₁s- for θεός (theós)). My sources claim them to be quite distinct. -Atelaes λάλει ἐμοί 12:05, 16 May 2012 (UTC)
Maybe the common root is older than that and lost? That doesn't qualify as grounds to denounce the appearant relation, anyhow. I'll reword that sentence now --
I would like to point out that one of the five Latin League nations was the Veneti, amber traders from the Baltic. Those veneti (I reference information from the Lithuanian History museum in Klaipeda here) Were in turn migrants from Greece, with greek dress, customs, and IndoEuropean language. Taat does not prove origin, but it provides a mechanism by which Theos and Deus may be related.2600:1003:B119:7649:5424:A94D:569E:4208 16:42, 26 December 2014 (UTC)mjr
θεός came from PIE meaning maker. Deus came from PIE meaning sky, heaven, shine. Deus is cognate with Spanish día, but not with English day. θεός is cognate with English do. θεός and deus are false cognates, their PIE roots are not related to one another. —Stephen (Talk) 06:45, 6 April 2014 (UTC)

Catholic encyclopedia[edit]

The current citation on the current content of the page is ALSO garbage. It in no way supports the claims that the words are not related. Page 1 is "introduction to the comparatve method and the indo-european family" If anything, It actually supports the fact that these two words ARE INDEED related. But if we are desperate for a resource, the here is a quote from ... The Catholic Encyclopedia the first portion (not bold) is about the word "god"

"The root-meaning of the name (from Gothic root gheu; Sanskrit hub or emu, "to invoke or to sacrifice to") is either "the one invoked" or "the one sacrificed to." From different Indo-Germanic roots (div, "to shine" or "give light"; thes in thessasthai "to implore") come the Indo-Iranian deva, Sanskrit dyaus (gen. divas), Latin deus, Greek theos, Irish and Gaelic dia, all of which are generic names; also Greek Zeus (gen. Dios, Latin Jupiter (jovpater), Old Teutonic Tiu or Tiw (surviving in Tuesday), Latin Janus, Diana, and other proper names of pagan deities. The common name most widely used in Semitic occurs as 'el in Hebrew, 'ilu in Babylonian, 'ilah in Arabic, etc.; and though scholars are not agreed on the point, the root-meaning most probably is "the strong or mighty one.""

The main issue with sayin they aren't etymologically related is that the oldest surviving texts referencing the two usages play out like this... The Greek Bible wrote "theos" and then the Romans translated that EXACT word the "DEUS"... THAT is a relationship. Etymology goes both ways. It's not about which words are the precursors to later words, it's about the history of the usage of words & if one word gets directly transliterated to another in the same language family, THAT makes them cognate.

Okay, well, maybe not exactly... Let me give you an example. IN modern english we use the word "and" now, if by chance we as a society decided to stop using the word "and" and begin using the greek word "kai", then the etymology section for the word "and" would say something like "replaced by kai"... understood? Lostubes (talk) 11:46, 10 July 2016 (UTC)

You do know what "cognate" means, right? It means etymologically related, having a common origin (literally "born together" in Latin). Having similar or even identical meanings is no indication at all. see is not a cognate of voir, even though they have the same meaning. The English cognate to voir is in fact wit. —CodeCat 13:08, 10 July 2016 (UTC)
yes, exactly what I pointed out with the quote from the encyclica. from the two indogermanic roots,div & thes, an entire gamut of generic god names arises in the Mediterranean region... Lostubes (talk) 19:10, 10 July 2016 (UTC)

I guess it would be prudent to point out that thessasthai & theasthai are also related, the definition "to implore" is given, but other important translations cognate to the first given are "behold" & "theater" Hence the cintextual relationship between the two indo Germanic roots and subsequent evolution thereof. Any way yeah Lostubes (talk) 19:25, 10 July 2016 (UTC)

Herman Bavinck[edit]

From Definition of God's proper names by Herman Bavinck.

"Formerly the Greek word theos was held to be dreived from tithenai, theein, theasthai. At present some philologists connect it with Zeus, Dios, Jupiter, Deus, Diana, Juno, Dio, Dieu. So interpreted it would be identical with the Sanskrit "deva," the shining heaven, from "div," to shine. Others, however deny all etymological connection between the Greek theos and the Latin Deus and connect the former with the root thes in thessasthai to desire, to invoke."

—This comment was unsigned.

the harm[edit]

The reason this has been bugging me for years is this is contraverse information and some charlatans keep removing it.Lostubes (talk) 19:36, 10 July 2016 (UTC)

So, I'd point out again that the Catholic Encyclopedia does not constitute a reliable Proto-Indo-European source. The etymological practices of religious scholarship has always been mired in conflicts of interests, particularly a wish to syncretize disparate linguistic data to provide a unified and overly simplified explanation to demonstrate the presence of intelligent design. Furthermore, you still have not understood the main sticking point here: the meaning of there words are irrelevant; what is necessary is that they demonstrably come from the same PIE word, where demonstration requires adherence to a massive literature describing the sound changes between PIE and its descendant languages. PIE *dʰ- gives Greek θ and Latin f-. θεός does have several Latin cognates, specifically fēriae, fēstum, and fānum. Latin deus also has cognates of the underlying root (*dyew-) in Greek, namely Ζεύς (Zeús). There is no number of quotes you can provide that will change the underlying sound laws that gave these two disparate words. —JohnC5 19:45, 10 July 2016 (UTC)
your kidding... right? —This comment was unsigned.

Charles George Herbermann[edit]

(8 December 1840 – 24 August 1916) was born in Saerbeck near Münster, Westphalia, Prussia, the son of George Herbermann and Elizabeth Stipp.[1] He arrived in the United States in 1851, and seven years later graduated at College of St. Francis Xavier, New York City. He was appointed professor of Latin language and Literature (1869-1914) and librarian (1873-1914) at the College of the City of New York. For more than 50 years, he was immersed amidst various issues involved with Catholicism. He was president of the Catholic Club (1874–75) and of the United States Catholic Historical Society (1898-13). He became editor in chief of the Catholic Encyclopedia in 1905.[1] He translated Torfason's History of Vinland and wrote Business Life in Ancient Rome (1880).

I don't believe you realize that this debate is hundreds of years old.

Anyway I'm done here, enjoy your malignant encyclopedia website.Lostubes (talk) 20:02, 10 July 2016 (UTC)

Modern citations from respected Proto-Indo-Europeanists. Just because a debate has been around for a long time does not mean the older sources have any relevance. The majority of the history of medical literature is completely irrelevant because the theories on which it was based (humors, homunculi, etc.) were completely wrong. So too are the writings of older philologists often wrong and irrelevant to modern linguistics. Also, we are a malignant dictionary, not a malignant encyclopedia. —JohnC5 20:13, 10 July 2016 (UTC)

Established etymology[edit]

Chantraine, in his entry on 'θεός', states bluntly: 'Etymologie: Inconnue.' He then presents a few candidates, including the one given here as being established, concluding: 'finalement l'ensemble reste incertain'.

Likewise in Frisk: 'Nicht sicher erklärt.', and the same objection to linking it with *dʰ(e)h₁s: 'ē : ĕ bleibt noch zu erklären'.

Does anyone have sources stating that *dʰ(e)h₁s is now the established etymon?

Beekes says that that etymology "seems to be generally accepted". --Ivan Štambuk (talk) 21:35, 19 November 2013 (UTC)

deus dei di & uncertainty[edit]

The etymology section had previously stated that deus is not related to theos. I found this annoying as the vulgate used the deus dei di, etc. to translate the greek forms theos theon etc. So, in truth they are obviously comparable in the least.

The similarities aren't just superficial... they actually share meaning. In case you didn't notice that part...

Furthermore, the origins of the theos are UNKNOWN according to Strong. There is no evidence supporting the claim that these two words are in and of themselves entirely isolated from one another, in fact the evidence seems to trend in the other direction.

one word gets used in place of a previous word and the previous word is of unknown origin... these two words are inherently related in the context that the newer (latin deus) is the closest form of the word with any clear origin that we have...

that's not to say that it's correct to assume that they are the same word... but saying they aren't comparable... you might as well say that theos actually doesn't mean anything at all.

They cannot be related as Greek th- corresponds to Latin f-, while Greek d- corresponds to Latin d-. The discrepancy in the initial consonant makes common origin impossible. —CodeCat 00:41, 2 August 2014 (UTC)
(edit conflict) First of all, this is in the etymology section, so the only relationship that counts is in the derivation of the words. Of course θεός and deus have the same meaning, but that's irrelevant to the etymology. You mention Strong (I'm assuming you're referring to w:Strong's Concordance): it was compiled over a century ago by a theologian with no training in Indo-European historical linguistics. The evidence for the origin of both terms is pretty solid if you know about reconstruction of Indo-European roots, and it's not hard at all to tell that they come from completely different origins within Proto-Indo-European, and have just converged to have the same meaning. It's rare, but it happens. Chuck Entz (talk) 00:54, 2 August 2014 (UTC)


To pick up the old discussion, I'll replace the word "completely", because I miss evidence apart from the PIE-roots that theos and deus aren't in fact somehow related to a word older than that. To address the comment further down,

"The evidence for the origin of both terms is pretty solid if you know about reconstruction of Indo-European roots, and it's not hard at all to tell that they come from completely different origins within Proto-Indo-European",

I call that Original Research. *dʰeh₁- and *dyew- look strikingly similar and possibly sound similar. What's so obvious that the PIE-roots must have entirely different etymology? Who could even say that, when the reconstructions are unsure?

And the cited source (Fortson) does not actually proof anything, I'm reluctant to remove the Ref., though. It's hard to prove that something doesn't exist. So, either leave it out and let the roots speak for themselves, which would be a shame if it was correct after all, or hold the source to a higher standard if the point is contested. 01:53, 12 March 2017 (UTC)

No, I call that centuries of scholarship by people who are actually studying language- not religion, not underwater basketweaving. Insisting that it should be thrown out because you have a hunch- that's original research. There are very solid patterns in the distribution of sounds in Indo-European languages that have been well known for centuries, with a few well-documented exceptions that follow clear rules, such as w:Grassmann's law and w:Verner's law. I've taken classes in this as part of my linguistics degree, taught by people that have Wikipedia articles about them, and I'm not half as knowledgeable as some of the other people who have weighed in here to tell you you're wrong. The problem is that you're asking us to prove the validity of the whole field of Indo-European linguistics in a single talk page. Chuck Entz (talk) 06:10, 12 March 2017 (UTC)
"that's original research" - I didn't claim any positive results on my part. I doubt the linked source goes into detail besides saying the similar appearance was arbitrary. I didn't read the whole book, so take this with a grain of salt. The purpose of a source is to clarify (not to authorize, the cited first page does nothing but), maybe less if targeting a non-professional audience. As amateur I accept that I shouldn't editorialize. I'm asking honest questions, albeit honestly naive. But I wish to remain skeptic, as I assume this is an area of ongoing research.
"a few well-documented exceptions" - Is that in response to "reconstructions are [generally] unsure"? I suppose a few thousand years ago, exceptions were the norm. Even today local exceptions are far more than few, though repetitive. I don't have to prove the connection to raise doubt, do I?
Are deywós-derivatives missing in classic greek? Wouldn't that at least imply synonymy for some time before the loss of one stem? Then theos could be a Back-formation to deus, vice-versa to the hint down-page, that the formation of deus could be informed via Phrygian Language.
Ζεύς was traced back to *dyew- for example, with the Laconian variety Δεύς (Deús) and Boeotian variety Θιός (Thiós). It's hard to believe that wouldn't be an influential parallel. 07:02, 13 March 2017 (UTC)
"Basketweaving" - are you assuming good faith? I'm sorry that my doubt seems offensive to you. Of course I'm provocative on purpose, but I think I remained reasonably objective. You should see the wild theories that I didn't just dump here. After all the only solid claim was, PIE-roots aren't the beginning of language. E.g. If the disconnect could be clarified to well before any Indo-European expansion, my curiosity would be satisfied as well. I know that's a lot to ask. 06:45, 13 March 2017 (UTC)
*dʰeh₁- and *dyew- only look vaguely similar. But it is a deceptive similarity. They share the letters d and e in the spelling, but they do not have the same phonemes. One has two phonemes, *dʰ and *h₁, and the other has three phonemes, *d, *y, and *w. The e is just an ablaut vowel; the roots are really *dʰ_h₁- and *dy_w- with a hole where a vowel can be inserted or not. So really the roots are not similar at all: they do not share a single phoneme. — Eru·tuon 07:07, 12 March 2017 (UTC)
"they do not share a single phoneme." - That's what I mean is quite unsure: those phonemes are in broadly fuzzy categories, as far as I have understood, the roots are radicals, not complete words.
What on earth do you mean by "broadly fuzzy"? The very definition of a phoneme is that it stands in contradistinction with the other phonemes. You would not say that that the words pat and bad are at all the same word despite the fact that they share similar phonetic characteristics for both consonants (And to be clear, pat and bad are far closer than *dʰeh₁- and *dyew- are). Within a single language, words do not just suddenly change their pronunciation in a vacuum. —JohnC5 14:07, 13 March 2017 (UTC)
Is my assumption wrong about reconstructions mashing-up different local varieties? The phonetics are explicitly not clear, and probably varied wildly while and more importantly before words were written down frequently enough. If words wouldn't suddenly change with each new generation of speakers, if ever so slightly, they wouldn't change at all. 22:32, 13 March 2017 (UTC)
I suppose, the "-" as suffix is a place holder for inflection and composition; Then it could just repeatedly have grown and lost phonemes. The root-radicals are so short, the w:Hamming Distance would be only five coarse phoneme changes. At that, I guess dʰ and d aren't all that different (like th and d, e.g. thunder vs donner) and also "There are many variations of the laryngeal theory". So for lack of a better understanding I assume there could be an intermediate form. You get the gist, just my 2 cent 06:45, 13 March 2017 (UTC)
The difference between this situation and thunder and Donner is that they are in different languages and each of them descended exactly as expected from Proto-Germanic. Your argument here is "I can jury-rig some half-baked theories that could provide another explanation but nothing that supports my own argument nor anything that is more compelling than the actual evidence." —JohnC5 14:07, 13 March 2017 (UTC)
"provide another explanation" - As I said, I don't think I have to prove the connection to be skeptic. Any Half-Baked theory is just asking for clarification. I mean, you raise the point on the page and the source doesn't directly hint at a paper or any comprehensive explanation either. I'm within my rights to ask and I am thankful for good answers. It was claimed I asked for the "validity of the whole field of Indo-European linguistics in a single talk page", that shows that not only by my own estimate would the disconnect have to prove exhaustively. The difference is, you say that was merely because you knew too much to say about it, not that you couldn't for lack of a definite theory. Is research in that area is completed because of surprisingly many good sources? I suppose the opposite, but it doesn't even matter because PIE is neither the start nor the end of the story.
"exactly 'as expected from Proto-Germanic.'" - God is in its own category, it can have its own rules and you might just not know them. Now I'm gonna look forward to you claiming you do. Then I just can't argue with that anymore. 22:41, 13 March 2017 (UTC)

Quote: "as far as I have understood, the roots are radicals, not complete words". Could you clarify what you mean by radical? In Semitic linguistics, radical refers to one of the consonants in the usually three-consonant roots. This meaning does not work in the context of Proto-Indo-European.

As to the Hamming distance: I was not aware of this concept. So in this case a Hamming distance of 5 basically means that you can remove all the phonemes of one root and replace it with the phonemes of the other. Okay, so what? You could remove all the phonemes of any root and replace it with the phonemes of another. You could convert the roots *dʰ_h₁ and *dy_w into any other: *bʰ_r, for instance. Does that mean that all roots are related?

I was not aware of the Boeotian form Θιός (Thiós). It is much more similar to θεός (theós). I am not sure if there is a theory as to how it developed, or if it even comes from the same origin as Attic-Ionic Ζεύς (Zeús). It could actually come from θεός (theós), because then its form would be a regular development by the Thessalian Aeolic sound change of εο (eo) to ιο (io). In any case, I am not sure how the Boeotian form relates to the question of PIE relatedness of the two words we are talking about. — Eru·tuon 23:10, 13 March 2017 (UTC)

No written record of original IE exists, so reconstructions are likely 'radically' different from the actual words and the writing might actually have been consonantal alphabets or even pictographs for all we know. I wasn't under the impression the research suggests a homogeneous language anyway, but a family that shared some radicals (in the sense of stripped roots) and nobody denied that, only that the phonemes of PIE-roots could be constructed precisely because it is assumed that they didn't differ too much. But you can't use the assumption to prove its own validity.
With regards to Hamming Distance, I expected you to see that d-t has a smaller distance than e.g. d-k and I assume the visual similarity of and d was kind of a hint in that direction. Is that just coincidence? I wasn't sure, so I just assumed a fixed distance of 1 for any deletion or insertion, for sake of the argument.
The mentioned relation of Θιός (Thiós) to θεός (theós) and Ζεύς (Zeús): The whole point of the argument is that visual similarity can be arbitrary. Yet you agree on the possible influence of *theos* on this formation. You argue a known rule for that transformation. So, do you know a rule that prohibits a common development of *dʰeh₁- and *dyew-? The linked source doesn't supply one (afaik), hence it's not qualified. The qualification should be the source of the authority, not the other way around. —This unsigned comment was added by (talk) at 17:42, 14 March 2017 (UTC).
Ok, so what is your proposed change again? I find it unlikely that it will be added, given that modern scholarly consensus is against your proposal, but please remind me exactly what you want. Also, why do you care? You've been harping on this for a while now, despite the evidenceof several highly reputable members of our community who have provided well regarded sources for their claims. —JohnC5 20:36, 14 March 2017 (UTC)
Where are those other sources that you mention, besides Fortson? 21:49, 14 March 2017 (UTC)
There are plenty of sources that do directly support this claim on the pages for *dʰéh₁s and *deywós. So, again, what are you proposing? —JohnC5 02:26, 15 March 2017 (UTC)
Those sources are long books. Can you give page numbers so that we can add these to this article? A first search for "god" in (Ringe) and the lemma for θεός in (Beekes) didn't yield any specific claim. Optimally a source would make that claim and immediately prove it. 11:01, 15 March 2017 (UTC)
All of those sources except Ringe list the specific page numbers referenced. I'll say that each of these sources directly shows the etymology of specific terms (either from *dʰéh₁s or *deywós). They do not specifically say θεός (theós) is not a cognate of deus because the cognacy of those two terms is literally not possible within the serious study of PIE. Sources do not have say something specifically for their content to directly support a claim. Beekes says Ζεύς (Zeús) comes from *deywós and elsewhere that θεός (theós) comes from *dʰéh₁s. Similarly De Vaan ascribes fēriae and deus respectively to these separate roots. This is how the evidence works. Regardless, it is not my duty to do research for you or teach you how scholarship works. Please answer my question: what is your proposed solution? Unless you have something else to offer, this argument is meaningly (not to say it isn't meaningless to begin with). —JohnC5 15:16, 15 March 2017 (UTC)
"'Sources do not have say something specifically for their content to directly support a claim.'" - I think they do have to. May I direct you attention to {w: quote from Carl Sagan} (at the end of the intro).
"'They do not specifically say θεός (theós) is not a cognate of deus because the cognacy of those two terms is literally not possible within the serious study of PIE.'" - Then why would you write it in the article? 16:38, 15 March 2017 (UTC)
We put it in the article because you keep trying to add this claim in. The only reason we are having this discussion is that you want something changed but cannot actually say what. I would gladly remove the remark in the etymology, except that this old myth about the etymology exists. It does not, however, appear in modern etymology resources because it is obviously false to people familiar with the topic. Modern medical text do not have to refute the theory of humours whenever a new article is published. Now either say what you want done or stop whining. —JohnC5 17:49, 15 March 2017 (UTC)
If unfounded claims are added to the article, feel free to remove them. If you could find NPOV sources or at least multiple sources that show the claim is generally accepted, the claim would have a leg to stand on. But as far as I'm aware, personal opinion is abundant in etymological research.
As it stands, the claim is not limited to the phonemes, although that appears to be your only argument. I'd settle to accept "*There is no indication that these IE forms are cognate*". 21:30, 15 March 2017 (UTC)
What on earth does "the claim is not limited to the phonemes" mean? All the sources mentioned are NPOV and do support the claim. Again, if you do not intend to read them and the relevant literature to understand them, then you do not possess the information necessary to evaluate these claims. That is, of course, fine, but you can't go around saying that they don't support our stated theory without understanding how historical phonology, historical semantics, and Indo-Europeanism. I'm afraid, as it stands, I will not support your proposal.
On a side note, why do you care so much? Would you consider yourself religious? Just curious. —JohnC5 03:33, 16 March 2017 (UTC)
I will address that post later. For now I'd like to know if you are ignoring on purpose that cognacy is not limited to PIE. Just curious. 14:43, 17 March 2017 (UTC)
Oh, of course there is cognacy in other language families, but the degree to which PIE has been studied means we know a great deal about the complex set of rules that operate within of PIE and its reflexes. Just be cause you know generally how etymology works does not mean you are qualified to judge specific findings of Indo-Europeanism. I, for instance, know very little about Sino-Tibetan or Uralic families and could not hope to correctly evaluate etymologies in those families without a good deal of study. —JohnC5 15:06, 17 March 2017 (UTC)
That is neither a yes nor a no. Can you with certainty deny any relation of precursors to the PIE-roots in question? 16:19, 17 March 2017 (UTC)
Oh, are you asking whether, before PIE, the roots *dʰeh₁- and *dyew- were related? Since they are far apart in both meaning ("give" and "sky") and phonology, it does not seem plausible in the slightest. Also, the time-depth of pre-PIE and lack of evidence makes such speculations meaningless. —JohnC5 16:58, 17 March 2017 (UTC)
No, I was asking for the precursors in pre-PIE. If those were meaningless, PIE wouldn't have developed in the first place. If you think pre-PIE is meaningless, that's your opinion, it certainly wasn't to the speakers of pre-PIE.
Please look at *gʰeb-, *dʰeh1, *deh3 and then at *dédeh₃ti (to be giving) and *dʰédʰeh₁ti (to be putting). Is this obvious similarity in meaning and reconstructed phonemes also arbitrary? 02:45, 18 March 2017 (UTC)
As I said, while an ancestor to PIE existed, we do not have the comparative material to make claims about such a language. Again, you have misunderstood. I said speculation on it is meaningless because we have no evidence to reconstruct it accurately. You have decided instead to misinterpret that statement either willfully or confusedly into meaning that I somehow was insulting the speakers of pre-PIE. Your lack of basic understanding of the principles of linguistics, scholarship, and argumentation is very disconcerting.
As to the tangential set of roots you have listed, what are you arguing? That these roots look somewhat similar (despite none of them sharing any phonemes with one another besides the default *-e-) and have similar meanings? These roots are indeed unrelated and arbitrary. It's like ask whether the verbs to pat and to bat must be related because they are similar and meaning and phonology (kind of). I'll give you a hint, it is arbitrary. There isn't the slightest evidence that *dʰeh₁- (to give) and *dyew- (sky) are related in PIE or before.
So let's recap:
  • According the most highly respected linguists in the fields of Indo-Europeanism, including area experts on Hellenic and Italic, θεός (theós) and deus are from different roots.
  • The roots *dʰeh₁- (to give) and *dyew- (sky) share no similarity in meaning, and no phonemes in common.
  • Phonemes, by there nature, contrast with one another and do not just interchange outside of the government of rules.
  • Without more comparative evidence from another language family, it is impossible to know very much at all about pre-PIE, so any argument made about such language stages is mere flight of fancy.
  • All the citations you have given are from sources that are older than the modern understanding of Indo-European or written by non-linguists, particularly theologians with a religious agenda.
  • There is no reputable modern linguist who supports the theory that θεός (theós) and deus are cognates (though, to the layman, they may look so).
  • Therefore, there is no reasonable evidence in favor of your proposal, whereas there is ample evidence saying that the two words are unrelated.
With that in mind, it is my intention to cease engaging in this discussion. Feel free to hoot, holler, rant, and rave to your heart's content on this topic, but the evidence and the study of linguistics will mercifully remain unaltered. —JohnC5 03:19, 18 March 2017 (UTC)
The discussion continued elsewhere (see the link which user sche added below), I just want to address the points for completeness sake, by simply repeating what I said here before:
  • I didn't contest the validity of the different reconstructed roots at all, in fact I used them here.
  • The roots *dʰéh₁s (god, sacred place) and *dyew- (heaven) do share similarity in meaning. Both phonemes *dʰ and *d *are* rather close, both are coronal stops, only one is *supposed* to be voiced and the other aspirated. Nobody even knows what the fricatives sounded like.
  • Again, that's hard to argue with because nobody can actually know with absolute certainty what the phonemes sounded like or how they were written. Especially in religious texts, the writing could inform the pronunciation as well. That's doesn't require phonologic rules. In many cases the rules might describe inherent probabilities for changes to occur, but still, you can't claim to have an exhaustive list of rules, nor do they preclude exceptions to the rule. In some cases, changes in meaning might even motivate the sound changes. And rules about sound changes within PIE are as well uncertain.
  • "Without more comparative evidence from another language family, it is impossible to know very much at all" - That's all I am saying. Well, not exactly, the comparative method might be prevalent because of its success, but what I was alluding to is called w:Internal reconstruction!
  • You are misrepresenting what I said, I didn't give any sources at all.
  • We are not only talking about the Greek and Latin words. That neither is a loanword from the other, that's already clear from the etymology. That's besides the point. The added paragraph implies more than that.
  • You are misrepresenting my claim again, I don't argue the words have to be related. Anyhow, the sources don't directly show any evidence, but they would have to to qualify as reliable. 20:01, 27 March 2017 (UTC)
You ask me for a phonological rule that prohibits something. Typically, rules describe what happens, not what doesn't happen. So I would counter: do you know of a phonological rule that would explain such a transformation? — Eru·tuon 23:57, 14 March 2017 (UTC)
No, I don't know the next thing about linguistics. I understand that my doubt therefore must sound kind of offensive. But I said before, I don't have to prove the connection to raise doubt, and that's especially true if you assume you can't exactly prove the opposite. That sounds as if I still wouldn't want to understand the posted arguments. But I got good answers so far. 15:46, 15 March 2017 (UTC)
What I was saying about the Boeotian form is that I doubt it really etymologically relates to Ζεύς (Zeús). It looks like it is simply the word θιός (thiós, god) being used for a particular god. I say that because I do not know of a sound change by which the sound ζ (z) would randomly transform to θ (th). But I'm speculating. — Eru·tuon 23:57, 14 March 2017 (UTC)
The kind of replacement you describe wouldn't require slow phoneme changes, so I likened that to *dʰéh₁s (sacred place) replacing *dyew_ (sky, heaven) for example. Or a hint that a connection was mistakenly assumed (not to say known) between descendants of the two roots in question, by some. But, please don't ask me for original research, as much as I'd like to go on. 15:46, 15 March 2017 (UTC)
I would urge you to read the statement that θεός (theós) is not related to deus as saying "to the best of our knowledge, the words are not related". That is, based on the generally accepted reconstructed forms, and the known sound changes between PIE and Latin or Greek, there is no way for the words to be related. Perhaps there is some possibility that the two forms are related in some way that is outside our current body of knowledge. I don't know how to prove a negative. — Eru·tuon 23:57, 14 March 2017 (UTC)
That reading was exactly how I edited it.
I noticed that I don't really know what etymological connection means. Maybe that's the root of the misunderstanding (no pun intended). 15:46, 15 March 2017 (UTC)
In the context of Latin and Ancient Greek forms, etymological connection refers to two different things: two or more terms being derived from the same Proto-Indo-European root, or being derived from an identical PIE form. Several examples: Latin genus and Ancient Greek γένος (génos) derive from the same PIE form, *ǵénh₁os, from the same root, *ǵenh₁-. The Ancient Greek verb γίγνομαι (gígnomai, to be born; to happen) is related to the noun γένος (génos), because they come from different PIE forms from the root *ǵenh₁-. And the Latin noun genus is related to the Ancient Greek verb γίγνομαι (gígnomai), because the two come from different forms from the same root. That's a decent sampling of the different ways in which two Indo-European words from separate family branches may be related.
You edited the statement to say "an etymological relation isn't known". I guess I see now what your intention was, but I don't think that wording was very clear. I do not have a good replacement wording, but I will continue to think about it. — Eru·tuon 06:05, 20 March 2017 (UTC)

If the roots were synonym in metaphorical usage or meaningfully connected in some other way, that would be of interest, too. I mean, the language reflects the understanding of the people. And I imagine room for interpretation, becaue the topic of religion is inherently interpretive (that's what I mean with different category). 15:46, 15 March 2017 (UTC)

Well, Ζεύς (Zeús) and θεός (theós) are both in the topical category Category:grc:Gods, because they relate to the same topic. That is generally the way connections in meaning are noted in Wiktionary. I am not sure how metaphorical connections or other connections would be relevant to the Etymology section in this case. — Eru·tuon 06:05, 20 March 2017 (UTC)