How to design a non-European grammar

[Chagen]

Heocg.

This one was trying to be European but the Dorishar sprachbund cockblocked it endlessly. Let's see how much:

Five points:
postnominal relative clauses with inflected, resumptive relative pronouns (e.g. English "who" vs. "whose")
a periphrastic perfect formed with 'have' plus a passive participle (e.g. English "I have said")
a preponderance of generalizing predicates to encode experiencers, i.e. experiencers appear as surface subjects in nominative case (e.g. English "I like music" instead of "Music pleases me")
a passive construction formed with a passive participle plus an intransitive copula-like verb (e.g. English "I am known")
dative external possessors (e.g. German "Die Mutter wusch dem Kind die Haare")
verbal negation with a negative indefinite (e.g. English "Nobody listened")
particle comparatives in comparisons of inequality (e.g. English "bigger than an elephant")
equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant")
subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous (no marking of subject in the first place)
differentiation between intensifiers and reflexive pronouns
no distinction between alienable (e.g. legal property) and inalienable (e.g. body part) possession
no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you")
no productive usage of reduplication
no marking of arguments other than the subject on the verb
obligatory plural marking and no distinction between plural and collective

Two points:
definite and indefinite articles
verb-initial order in yes/no questions
comparative inflection of adjectives (e.g. English bigger)
conjunction A, B and C
suppletivism in second vs. two
topic and focus expressed by intonation and word order, not particles or affixes
only one gerund, preference for finite subordinate clauses
specific "neither-nor" construction
predominantly suffixing inflectional morphology
nominative–accusative morphosyntactic alignment

Roughly 53% SAE in grammar.

[KathTheDragon]

There is clearly something very wrong with this test.

[Chagen]

There is clearly something very wrong with this test.

I don't see that many problems with it. It encapsulates SAE grammatical tendencies rather well. Some of the questions are confusing, though.

[Ser]

Confusing to the point a couple of them seem to say the opposite of what they mean!

[KathTheDragon]

There is clearly something very wrong with this test.

I don't see that many problems with it. It encapsulates SAE grammatical tendencies rather well. Some of the questions are confusing, though.

I mean in how points are given.

[WeepingElf]

Old Albic

Five points:
postnominal relative clauses with inflected, resumptive relative pronouns (e.g. English "who" vs. "whose")
a periphrastic perfect formed with 'have' plus a passive participle (e.g. English "I have said")
a preponderance of generalizing predicates to encode experiencers, i.e. experiencers appear as surface subjects in nominative case (e.g. English "I like music" instead of "Music pleases me")
a passive construction formed with a passive participle plus an intransitive copula-like verb (e.g. English "I am known")
dative external possessors (e.g. German "Die Mutter wusch dem Kind die Haare")
verbal negation with a negative indefinite (e.g. English "Nobody listened")
particle comparatives in comparisons of inequality (e.g. English "bigger than an elephant")
equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant")
subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous
differentiation between intensifiers and reflexive pronouns
no distinction between alienable (e.g. legal property) and inalienable (e.g. body part) possession
no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you")
no productive usage of reduplication
no marking of arguments other than the subject on the verb
obligatory plural marking and no distinction between plural and collective

Two points:
definite and indefinite articles
verb-initial order in yes/no questions
comparative inflection of adjectives (e.g. English bigger)
conjunction A, B and C
suppletivism in second vs. two
topic and focus expressed by intonation and word order, not particles or affixes
only one gerund, preference for finite subordinate clauses
specific "neither-nor" construction
predominantly suffixing inflectional morphology
nominative–accusative morphosyntactic alignment

31.5% SAE in grammar, unless I miscounted.

[WeepingElf]

Roman Germanech

Five points:
postnominal relative clauses with inflected, resumptive relative pronouns (e.g. English "who" vs. "whose")
a periphrastic perfect formed with 'have' plus a passive participle (e.g. English "I have said")
a preponderance of generalizing predicates to encode experiencers, i.e. experiencers appear as surface subjects in nominative case (e.g. English "I like music" instead of "Music pleases me")
a passive construction formed with a passive participle plus an intransitive copula-like verb (e.g. English "I am known")
dative external possessors (e.g. German "Die Mutter wusch dem Kind die Haare")
verbal negation with a negative indefinite (e.g. English "Nobody listened")
particle comparatives in comparisons of inequality (e.g. English "bigger than an elephant")
equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant")
subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous
differentiation between intensifiers and reflexive pronouns
no distinction between alienable (e.g. legal property) and inalienable (e.g. body part) possession
no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you")
no productive usage of reduplication
no marking of arguments other than the subject on the verb
obligatory plural marking and no distinction between plural and collective

Two points:
definite and indefinite articles
verb-initial order in yes/no questions
comparative inflection of adjectives (e.g. English bigger)
conjunction A, B and C
suppletivism in second vs. two
topic and focus expressed by intonation and word order, not particles or affixes
only one gerund, preference for finite subordinate clauses
specific "neither-nor" construction
predominantly suffixing inflectional morphology
nominative–accusative morphosyntactic alignment

100% SAE in grammar, unless I miscounted.

[Nortaneous]

Confusing to the point a couple of them seem to say the opposite of what they mean!

rewrite it then, this is decidedly not my field

I'm the kind of person that looks at paperwork and goes "But what do they mean?", so bear with me. I have some questions.

phrasal adverbs (e.g. English already, still, not yet);

Anyone know what the hell that actually means? How are these "phrasal"?

This one was dropped from the list, so it probably doesn't matter anymore.

that is exactly why i dropped it.

obligatory plural marking and no distinction between plural and collective

I'm pretty sure you meant "no dictinction between singular and collective". ...R-right? Latin wouldn't qualify presumably, because of the existence of certain pluralia tantum that are basically collective nouns that only appear in plural-like forms (e.g. arma 'weapons', castra 'military camp'). But English, Spanish and French would qualify, since AFAIK there's no such nouns in these languages.

'no distinct collective' maybe?

[Rhetorica]

Lilitika:

Five points:
postnominal relative clauses with inflected, resumptive relative pronouns (e.g. English "who" vs. "whose")
a periphrastic perfect formed with 'have' plus a passive participle (e.g. English "I have said")
a preponderance of generalizing predicates to encode experiencers, i.e. experiencers appear as surface subjects in nominative case (e.g. English "I like music" instead of "Music pleases me")
a passive construction formed with a passive participle plus an intransitive copula-like verb (e.g. English "I am known")
dative external possessors (e.g. German "Die Mutter wusch dem Kind die Haare")
verbal negation with a negative indefinite (e.g. English "Nobody listened")
particle comparatives in comparisons of inequality (e.g. English "bigger than an elephant")
equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant")
subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous (one or the other depending on dialect)
differentiation between intensifiers and reflexive pronouns (just the adverb for 'alone')
no distinction between alienable (e.g. legal property) and inalienable (e.g. body part) possession
no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you")
no productive usage of reduplication (rare but productive, as an intensifier)
no marking of arguments other than the subject on the verb
obligatory plural marking and no distinction between plural and collective (whole-class infix -in- used with singular)

Two points:
definite and indefinite articles
verb-initial order in yes/no questions
comparative inflection of adjectives (e.g. English bigger)
conjunction A, B and C
suppletivism in second vs. two
topic and focus expressed by intonation and word order, not particles or affixes
only one gerund, preference for finite subordinate clauses
specific "neither-nor" construction
predominantly suffixing inflectional morphology
nominative–accusative morphosyntactic alignment

52/95=55% European... I really expected higher, but that number fits well with the backstory. Not bad for not trying, I guess.

[Yaali Annar]

Indonesian:

• Postnominal relative clauses with inflected, resumptive relative pronouns (e.g. English "who" vs. "whose") - 0

• a periphrastic perfect formed with 'have' plus a passive participle (e.g. English "I have said") - 0

• a preponderance of generalizing predicates to encode experiencers, i.e. experiencers appear as surface subjects in nominative case (e.g. English "I like music" instead of "Music pleases me") - 5

• a passive construction formed with a passive participle plus an intransitive copula-like verb (e.g. English "I am known") - 0

• dative external possessors (e.g. German "Die Mutter wusch dem Kind die Haare") - 0

• verbal negation with a negative indefinite (e.g. English "Nobody listened") - 5

• particle comparatives in comparisons of inequality (e.g. English "bigger than an elephant") - 5

• equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant") - 5

• subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous - 0

• differentiation between intensifiers and reflexive pronouns - 0

• no distinction between alienable (e.g. legal property) and inalienable (e.g. body part) possession - 5

• no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you") - 0

• no productive usage of reduplication - 0

• no marking of arguments other than the subject on the verb - 0

• obligatory plural marking and no distinction between plural and collective - 0

Two points:

• definite and indefinite articles - 0

• verb-initial order in yes/no questions - 0

• comparative inflection of adjectives (e.g. English bigger) - 0

• conjunction A, B and C - 2

• suppletivism in second vs. two - 0

• topic and focus expressed by intonation and word order, not particles or affixes - 2

• only one gerund, preference for finite subordinate clauses - 0

• specific "neither-nor" construction - 0

• predominantly suffixing inflectional morphology - 0

• nominative–accusative morphosyntactic alignment - 2

31%

[Chagen]

I have a feeling that a variant of this test for classical Eurolangs (Latin, Sanskrit, Ancient Greek, etc.) would be useful, as they have a lot of differences with modern eurolangs--lots of participles, tons of cases, extensive conjugation, articles (except in Ancient Greek, but apparently some earlier varieties didn't have them), requiring several principal parts to allow for full conjugation for a verb (though this is debatable given that many Germanic languages today need three principal parts for strong verbs), having many conjugational classes (Latin had 4...and-a-half-ish, don't know about Greek, Sanskrit had TEN), marking of passive on the verb, deponent verbs with middle voice-meanings that look like passive verbs (though from what I can tell Sanskrit atmanepada was a MIDDLE that could optionally be a passive--maybe "mediopassive-marking, some verbs only show up in passive marking but have active/middle meanings), some way to mark the dual (though it was pretty much dead even in the classical era), extensive use of irrealis for subordinate clasues...

Also, given that a ton of conlangers use classsical eurolangs as inspiration for their conlangs (usually by just ripping them off...I'm guilty of that too), it might be useful.

[Rhetorica]

English:
...
equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant")[/color](?)

If I'm not mistaken, the French translates to "big as an elephant," which English definitely lets you do.

[hwhatting]

obligatory plural marking and no distinction between plural and collective

I'm pretty sure you meant "no dictinction between singular and collective". ...R-right? Latin wouldn't qualify presumably, because of the existence of certain pluralia tantum that are basically collective nouns that only appear in plural-like forms (e.g. arma 'weapons', castra 'military camp'). But English, Spanish and French would qualify, since AFAIK there's no such nouns in these languages.

English has them as well (e.g. glasses (in the sense of "seeing aid"), scissors, trousers). But the distinction meant here is, I assume, a formal distinction between singular, collective, and plural, which most (all?) European languages don't have.

[ObsequiousNewt]

obligatory plural marking and no distinction between plural and collective

I'm pretty sure you meant "no dictinction between singular and collective". ...R-right? Latin wouldn't qualify presumably, because of the existence of certain pluralia tantum that are basically collective nouns that only appear in plural-like forms (e.g. arma 'weapons', castra 'military camp'). But English, Spanish and French would qualify, since AFAIK there's no such nouns in these languages.

English has them as well (e.g. glasses (in the sense of "seeing aid"), scissors, trousers). But the distinction meant here is, I assume, a formal distinction between singular, collective, and plural, which most (all?) European languages don't have.

Except in rare instances like person/people, which can hardly be called a singulative/collective distinction anyway. Are there any examples of IE languages with a productive (or previously productive) collective?

[clawgrip]

obligatory plural marking and no distinction between plural and collective

I'm pretty sure you meant "no dictinction between singular and collective". ...R-right? Latin wouldn't qualify presumably, because of the existence of certain pluralia tantum that are basically collective nouns that only appear in plural-like forms (e.g. arma 'weapons', castra 'military camp'). But English, Spanish and French would qualify, since AFAIK there's no such nouns in these languages.

English has them as well (e.g. glasses (in the sense of "seeing aid"), scissors, trousers). But the distinction meant here is, I assume, a formal distinction between singular, collective, and plural, which most (all?) European languages don't have.

These paired words, along with words like clothes, pyjamas, bangs (hair), (the) odds (of something happening), etc. that have no singular form are called plurale tantum, not collectives. Collective nouns are singular nouns that refer to a group, e.g. group, cluster, colony, team, everyone, etc.

The question is confusing because collective is a type of noun, but plural is an inflection.

[clawgrip]

Attempting to fix the questions:

- postnominal relative clauses with inflected, resumptive relative pronouns (e.g. English "who" vs. "whose")
- postnominal relative clauses with inflected relative pronouns (e.g. English "who" vs. "whose")
removed "resumptive" because these are not resumptive

- a periphrastic perfect formed with 'have' plus a passive participle (e.g. English "I have said")
no problem

- a preponderance of generalizing predicates to encode experiencers, i.e. experiencers appear as surface subjects in nominative case (e.g. English "I like music" instead of "Music pleases me")
no problem

- a passive construction formed with a passive participle plus an intransitive copula-like verb (e.g. English "I am known")
no problem

- dative external possessors (e.g. German "Die Mutter wusch dem Kind die Haare")
no problem

- verbal negation with a negative indefinite (e.g. English "Nobody listened")
no problem

- particle comparatives in comparisons of inequality (e.g. English "bigger than an elephant")
I don't have enough knowledge of comparisons in different languages to suggest a change. I know Chinese and Japanese also use particles. What's the alternative? A verb? "big exceed elephant"?

- equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant")
- equative constructions based on adverbial-relative clause structures (e.g. Spanish "grande como un elefante")
I understand now what this question means. Whoever wrote that example seems not to have understood the difference between "comme" and "comment" in French. I suggest Spanish as a better example because como (like) and cómo (how) are the same word and differentiated only in orthography.

- subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous
no problem

- differentiation between intensifiers and reflexive pronouns
no problem

- no distinction between alienable (e.g. legal property) and inalienable (e.g. body part) possession
no problem

- no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you")
- number distinction in pronouns, but no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you")
my guess is that a failure to distinguish number in pronouns at all is a decidedly non-SAE feature, so this should be clarified to exclude languages that do not differentiate number at all.

- no productive usage of reduplication
no problem

- subject marking on the verb but no other argument marking
no problem

- obligatory plural marking and no distinction between plural and collective
again, it depends on what exactly "collective" means. Will have to think about this one.

- grammatical sex marking
no problem

Two points:
- definite and indefinite articles
- obligatory definite and indefinite articles/suffixes
I think the idea here is that the language explicitly and obligatorily marks both definite and indefinite nouns, so this excludes Arabic for example, which has a definite article but lacks an indefinite article (right?), or Japanese, which lacks a definite article but has an optional indefinite (but specific) "article" (actually a one-word relative clause).

- verb-initial order in yes/no questions (should this count if the language is always verb-initial?)
- non-initial verb order in statements but verb-initial order in yes/no questions
I don't think it should include languages that are always verb initial, because I think the main point of this question is the swapping of the verb and subject to form questions.

- comparative inflection of adjectives (e.g. English bigger)
no problem

- conjunction A, B and C
no problem

- SVO or V2 word order
no problem, but it might help to specify clearly what they mean by V2, which I assume is forcing the second element of the sentence to be some kind of verb, like German (right?).

- topic and focus expressed by intonation and word order, not particles or affixes
no problem

- only one gerund, preference for finite subordinate clauses
not sure. These seem to be two different ideas merged into one.

- specific "neither-nor" construction
no problem

- predominantly suffixing inflectional morphology
no problem

- nominative–accusative morphosyntactic alignment
no problem

Any thoughts?

[Rhetorica]

V2 word order is what you think it is; cf. Wikipedia.

I'm not sure what's up with the collective-vs-plural thing, either; it's my understanding that collective is normally contrasted with singulative, i.e. when the unmarked form is plural instead of singular. I'm guessing the intended meaning was the presence of a dual number, or perhaps some similar number used with complete sets (such as "a pair of pants" or "a shelf of books").

[Hallow XIII]

Dead IE languages liked to use cases for comparison. Cf. the classic multo maior fratre statura (much-ABL bigger brother-ABL height-ABL): "by much taller than (his) brother". Ancient Greek did this with the Genitive. Certainly this is also found elsewhere?

[clawgrip]

I think we need to just scrap the comparative one. While it may be true that SAE languages tend to use particles for comparisons, I don't think this is exclusive enough to SAE to serve any useful purpose here. After all, this thread is about how to design a non-European grammar, not how to design a European grammar.

[hwhatting]

Dead IE languages liked to use cases for comparison. Cf. the classic multo maior fratre statura (much-ABL bigger brother-ABL height-ABL): "by much taller than (his) brother". Ancient Greek did this with the Genitive. Certainly this is also found elsewhere?

Russian still does this, as does Polish - in both, the genitive is used. Both languages also have the option of using a particle (чем in Russian, niż in Polish), with the case of the comparandum determined by its role in the sentence (чем in Russian, niż in Polish).

[Yng]

You can also use a preposition, as in Arabic:

أنا أكبر من أختي
ʾana akbaru min ʾuxt-ī
I bigger from sister=my
I'm older than my sister.

When the comparand is a clause you have to attach a nominalising -mā to min:

أنا أتكلم أسرع مما يتكلم
ʾana atakallamu ʾasraʿa mimmā yatakallamu
I speak faster than-NOM he_speaks
I speak faster than he does

I'm pretty sure with some adjectives you can also use the genitive:

أنا أكبره
ʾana akbaru-hu
I bigger=his
I am older than him

But I might have imagined this.

[Chagen]

There's also "exceed" comparisons. Maybe using a dative as well? I have Heoch do both:

Fah stut knaū br.
1SG.NOM that-DAT.SG fast be

(there is no comparative inflection the adjective here as the use of a dative implies a comparison. Alone, however [eg."I'm faster"] there is one: "Fah knaoyyū br")

Fah stuwa knaut ęw
1SG.NOM that-ACC.SG fast-DAT exceed

The second is considered highly archaic (but in Heocg speaking archaically is considered badass/cool so...)

[cromulant]

Gac

Five points:
postnominal relative clauses with inflected, resumptive relative pronouns (e.g. English "who" vs. "whose")
a periphrastic perfect formed with 'have' plus a passive participle (e.g. English "I have said")
a preponderance of generalizing predicates to encode experiencers, i.e. experiencers appear as surface subjects in nominative case (e.g. English "I like music" instead of "Music pleases me") it can do it either way
a passive construction formed with a passive participle plus an intransitive copula-like verb (e.g. English "I am known")
dative external possessors (e.g. German "Die Mutter wusch dem Kind die Haare")
verbal negation with a negative indefinite (e.g. English "Nobody listened")
particle comparatives in comparisons of inequality (e.g. English "bigger than an elephant")
equative constructions based on adverbial-relative clause structures (e.g. French "grand comme un élephant")
subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous (no marking of subject in the first place)
differentiation between intensifiers and reflexive pronouns
no distinction between alienable (e.g. legal property) and inalienable (e.g. body part) possession
no distinction between inclusive and exclusive first-person plural pronouns ("we and you" vs."we and not you")
no productive usage of reduplication
no marking of arguments other than the subject on the verb
obligatory plural marking and no distinction between plural and collective
grammatical sex-marking

Two points:
definite and indefinite articles
verb-initial order in yes/no questions
comparative inflection of adjectives (e.g. English bigger)
conjunction A, B and C
suppletivism in second vs. two
topic and focus expressed by intonation and word order, not particles or affixes
only one gerund, preference for finite subordinate clauses
specific "neither-nor" construction
predominantly suffixing inflectional morphology
nominative–accusative morphosyntactic alignment

21%

[cromulant]

Can we discuss the elephant in the corner?

grammatical sex-marking

People are taking two different versions of the test, that appear to differ solely in whether they have this question or not. I chose to include it as it brings the total number of points to 100 instead of 95.

[Basilius]

- subject person affixes as strict agreement markers, i.e. the verb is inflected for person and number of the subject, but subject pronouns may not be dropped even when this would be unambiguous
no problem

But isn't this a minority option in Europe? Most Romance, most Slavic, Greek etc. drop pronouns rather freely.

Then it boils down to this:

- subject marking on the verb but no other argument marking
no problem

Also...

- grammatical sex marking
no problem

IMO this needs some elaboration (in which sense English he/she is *grammatical*). "Grammatical gender" in general isn't precisely about "sex" (in Europe, at any rate).

- verb-initial order in yes/no questions (should this count if the language is always verb-initial?)
- non-initial verb order in statements but verb-initial order in yes/no questions
I don't think it should include languages that are always verb initial, because I think the main point of this question is the swapping of the verb and subject to form questions.

Good correction, but... a lot of languages in Europe don't *ban* verb-initial WO in declarative sentences. So I propose something like Mostly verb-initial WO in yes/no questions, with verb-initial WO not prevailing in declarative sentences.

- SVO or V2 word order
no problem, but it might help to specify clearly what they mean by V2, which I assume is forcing the second element of the sentence to be some kind of verb, like German (right?).

I have troubles with this, because "SVO" and "V2" are about different notions.

"V2" is about a set of restrictions which, combined, disallow placing the V to any other position in main declarative clause (or disallow placing it further to the right while permitting V1, e. g. in Icelandic and, as a marked option, in e. g. German). These restrictions don't usually tell you what to put before the verb, so a V2 language may permit XVSO or XVOS (besides SVO and OVS). In Europe, only Germanic languages are V2 (but English isn't).

In contrast, "SVO" is not usually understood in terms of restrictions (in particular, WALS does not use this label this way). When someone says that a language is "SVO", it doesn't usually mean that anything else is disallowed (or restricted to subordinate clauses, or non-declarative sentences etc.), for in this sense, there are, like, only two "SVO" languages in Europe (English and French). Normally labelling a language "SVO" means that SVO is the statistically prevailing WO, and there is usually some looseness in specific criteria (for example, WALS wants the most frequent WO to be at least twice as frequent as the next-most-frequent WO in order to be considered prevailing, else WALS will use the label "no predominant WO"; granted, this looks rather arbitrary).

"Statistically prevailing SVO" is indeed most common in Europe, but then it's not a very strong marker for it's also second-commonest globally.

- only one gerund, preference for finite subordinate clauses
not sure. These seem to be two different ideas merged into one.

To me too. Also, does "having come" count, and why if not?