My new subset of the study of grammar. Thoughts?

[AcademiaNut]

Over the years, as I've studied different languages and have written down patterns I've noticed, I began to notice that what I was doing was something I have not seen any language book do. What I wrote down were empirical rules based on statistics, and because rules of a language are considered grammar...

http://www.differencebetween.info/diffe ... vocabulary

...I now assume that my study was a subset of of the study of grammar. I would like to get people's thoughts on this type of work. For example: Has anybody else published something like this? Do you think it is useful? What should such a study be called?

Here are some examples of the patterns I've found that allow me to create new words...

(1)
Suppose you want the Latin word for "Internet." Obviously ancient Romans didn't have computers (only abacuses!) so they didn't have Internet, so they didn't have a word for Internet. However, it is possible to logically synthesize such a word. By looking at a number of Latin words that are translations of English words of the form interX, I found that the pattern was...

interX [English] = interX [Latin]

In this case I was lucky that the pattern was identical to English's pattern. The ancient Romans did have nets and networks, though, and a good Latin term for that concept is "rete." Now if you set X = net, the above equation produces...

Internet [English] = Interrete [Latin]

I've done this type of thing a lot with a lot of languages, for various reasons, but mostly for Latin since such words simply didn't exist in any available dictionary. Here is another example:

(2)
How would you say "printout" in Latin? By looking up words individually, we get...

to print [English] = imprimo/imprimere/impressi/impressum [Latin]
pressed [English] = presso/pressa/pressum [Latin]

...and by doing an empirical study of word translations for English to Latin I derived the pattern...

Xout [English] = exX [Latin]

...so substituting X = print, the equation produces...

printout [English] = eximpressum [Latin]

(3)
How would you say "cat food" in Portuguese?

food [English] = comida [Portuguese]
cat [English] = gato [Portuguese]

...and by doing an empirical study of word translations for English to Portuguese I derived the pattern...

XY [English] = Y de X [Portuguese]

...so substituting X = cat and Y = food, the equation produces....

catfood [English] = comida de gato [Portuguese]

In general, patterns in a language can usually be described by variables, then specific letters or words substituted into those variables to produce new words that are consistent with the empirical grammatical rules of the target language.

[Deinonysus]

It seems that you have stumbled onto the phenomenon of calques. It is very common (but nowhere near universal) for a language to take a word or phrase part by part from another language. A couple of examples off the top of my head is that the German word überleben is calqued from the Latin supervīvō (which English took a bit more directly as "survive") and the Modern Hebrew word for Apple, תפח אדמה (tapuakh adama, literally meaning "apple of the earth"), was calqued from the French pomme de terre.

You may be interested in the Anglish community on Reddit. They are focused on creating a version of English that is purged of Romance and Greco-Latin vocabulary, and they will often calque words from German, similar to what you are doing to adapt modern words into Latin. I believe the community was inspired by Uncleftish Beholding, a 1989 description of the chemical elements that used calques from German such as Waterstuff and Sourstuff, as well as coining original compound words such as "firststuff" (element) and "uncleft" (atom).

[DaveAgain]

Over the years, as I've studied different languages and have written down patterns I've noticed, I began to notice that what I was doing was something I have not seen any language book do. What I wrote down were empirical rules based on statistics, and because rules of a language are considered grammar...

http://www.differencebetween.info/diffe ... vocabulary

...I now assume that my study was a subset of of the study of grammar. I would like to get people's thoughts on this type of work. For example: Has anybody else published something like this?

Le Robert's Robert Micro dictionary has a section on suffixes/prefixes. I believe the Robert Brio analyses word formation/families to a greater depth, but I've not seen a copy.

[smallwhite]

I don’t know, but I love your username.

[Cainntear]

...I now assume that my study was a subset of of the study of grammar. I would like to get people's thoughts on this type of work. For example: Has anybody else published something like this? Do you think it is useful?

It is only useful to explain patterns that do occur, not to postulate ones that don't.

One way of looking at the structure of language is to split it into "grammar" and "idiom", where grammar is straightforward rules of combination, and idiom is where the combination of items generates a new meaning that cannot be predicted from the rules of grammar.

Most people think of "idioms" as meaning stock phrases, metaphors, analogies and cliches, but there's more to it than that.

Consider "I'm X years old". You may think of that as being straight application of grammatical rules, but in French they say "J'ai X ans" -- literally "I have X years". The grammar for both forms is possible in either language, but only one form has meaning (I.e. just as "I have 40 years" means nothing in English despite being grammatically possible, "Je suis 40 ans d'age" is grammatically possible but equally meaningless); the difference therefore is one of metaphor, hence idiom.

Both French and English see age as some kind of property, but English presents it as an attribute by using an adjectival form while French recognises its transience by treating it as a possession that will later be lost.

Here are some examples of the patterns I've found that allow me to create new words...

(1)
Suppose you want the Latin word for "Internet." Obviously ancient Romans didn't have computers (only abacuses!) so they didn't have Internet, so they didn't have a word for Internet. However, it is possible to logically synthesize such a word. By looking at a number of Latin words that are translations of English words of the form interX, I found that the pattern was...

interX [English] = interX [Latin]

In this case I was lucky that the pattern was identical to English's pattern. The ancient Romans did have nets and networks, though, and a good Latin term for that concept is "rete." Now if you set X = net, the above equation produces...

Internet [English] = Interrete [Latin]

And yet the Romance languages don't do this, and just use plain old rete, red etc. Why did they drop the inter?
Because interrete means "between-network", not "a network". The internet is a "network of networks", which makes it "a network". To translate it into Latin (which I don't speak) while retaining inter you'd really need to call it the "network internetwork", and then the question would be whether the majority of people would shorten it to the first element (network/net) or the second element (internetwork/internet). I suspect the latter, personally, based on the behaviour of the Romance language speakers.

The English coinage doesn't follow Latin rules closely enough to be translated directly.

How would you say "printout" in Latin? By looking up words individually, we get...

to print [English] = imprimo/imprimere/impressi/impressum [Latin]
pressed [English] = presso/pressa/pressum [Latin]

...and by doing an empirical study of word translations for English to Latin I derived the pattern...

Xout [English] = exX [Latin]

...so substituting X = print, the equation produces...

printout [English] = eximpressum [Latin]

Problem:
the "im-" in "impressum" is the Latin prefix "in-" as it occurs before a bilabial consonant, and this "in-" means "in".

The Latin "imprimere" means "to press in"; thus "eximprimere" means "to press in out".

The "out" in "print out" isn't describing direction -- rather it's the English idiom of "out" that exists in phrases such as "copy out". You wouldn't translate "copy out" to Latin with "ex-", because "copy" is a Latin word meaning "to copy" or "to copy out".

(3)
How would you say "cat food" in Portuguese?

food [English] = comida [Portuguese]
cat [English] = gato [Portuguese]

...and by doing an empirical study of word translations for English to Portuguese I derived the pattern...

XY [English] = Y de X [Portuguese]

...so substituting X = cat and Y = food, the equation produces....

catfood [English] = comida de gato [Portuguese]

In French, Italian and Spanish, cat food is "food for cats" -- respectively "nourriture pour chat","cibo per gatti" and "comida per gatos" -- so while I don't speak Portuguese, I'm pretty certain your conclusion is wrong.

Here we're on the boundaries of grammar vs idiom, because someone applying your rule for direct calquing the other way is going to say "food for cats" and will be understood, and people will probably understand the mistranslation the other way around, but there's an outside change that in Spain, where "comida" means both "food" and "meal", you might be misunderstood as asking for a "meal of cat".

Alternatively, the difference can be described grammatically as being about having "benefactive case marking", where "benefactive" means you have grammar showing "for whose benefit" something is.

In general, patterns in a language can usually be described by variables, then specific letters or words substituted into those variables to produce new words that are consistent with the empirical grammatical rules of the target language.

But the rules themselves are variables, and you have to select all of the right variables.

If you're a computer programmer, it's like language has first-class functions: functions are variables too.

[Cainntear]

Returning to "out"...

Consider that a "layout" is also known as a "plan". "Layout" has its roots in Germanic older forms of English, whereas "plan" is a more recent borrowing from Latin, probably via French.

The English "layout"/"to lay out" is a grammatical coinage based on the metaphor of placing things in position, the Latinate "plan" is a pure metaphor deriving from a word meaning "flat" or "flat and broad" (cf. "plane"/"planar" and "plate") -- when we draw a plan of a house, we make a flat representation of a 3D item.

Oh, and don't forget the third near-synonym: "blueprint". The grammatical coinage is clear, but the idiom/metaphor is being lost as we stop actually printing our blueprints in blue now that cyanotype is no longer cheaper than other printing technologies.

Three different metaphors, three different idioms, leading to three dramatically different terms with near-identical meanings.

[Ccaesar]

Sounds very logically, since in rete in Italian can also be used as "on the net/in the internet"

[AcademiaNut]

It seems that you have stumbled onto the phenomenon of calques.

Great information, thanks! I've never even heard of that word.

However, the equation method I'm using goes further than just word-for-word translation. For example, the conversion of the form "XY" to "Y de X" is more general than just borrowing root words: it deals with word arrangements. Also, my system applies to phonetics, not just to words or arrangements of words.

Example: In French, a word that ends in -c usually has the -c pronounced (as /k/), but sometimes the -c is silent. The empirical percentages turn out to be about 83% for pronounced, 17% for silent. Some examples of pronounced: lac /lak/, sac /sak/. Some examples of silent: tabac /taba/, estomac /Est)ma/. This could be notated as:

/Xc/ [French] = {83% /Xk/, 17% /X/} [SAMPA]

Example: In French, a word that ends in -ais virtually always has the -ais pronounced as /E/. Some examples: anglais /A~glE/, hollandais /)lA~dE/, japonais /Zap)nE/. This could be notated as:

/Xais/ [French] = /XE/ [SAMPA]

[AcademiaNut]

Le Robert's Robert Micro dictionary has a section on suffixes/prefixes. I believe the Robert Brio analyses word formation/families to a greater depth, but I've not seen a copy.

From the advertisement, it sounds like the prefixes and suffixes in that dictionary would just be for French to French. I have English dictionaries that have prefixes and suffixes, so that is not so unusual for a dictionary. However, I don't believe any of my French dictionaries have prefixes or suffixes, so that is a first for me. Thanks for the information.

[AcademiaNut]

I don’t know, but I love your username.

Thanks. I love your beak.