Ælis/Topicalisation

Ælis has no word classes, fixed word order, or word dividers. Since these aspects generally are rather important structural elements of languages, Ælis needs to have other methods of achieving consistency. One of these methods is topicalisation, also called topic marking or in this case function marking. What this means is that each word in a sentence is preceded by a root word which denotes the word's semantical function in the sentence. Fasten your seatbelt and assume brace position.

Functions
There are 8 functions in Ælis, divided into three categories.

Primary functions
Primary functions divide sentences into the equivalent of constituents (noun phrase, verb phrase, adverbial phrase, etc.). Since Ælis doesn't use spaces, the primary functions serve as a means of delineating different words within a sentence. There are 4 primary functions.

Topic

 * → Corresponding morpheme: hA [ha].

The topic describes an action or state. In full sentences, it will often be translated with a verb, though the topic itself has no form of conjugation, tense or mood. In essence, the topic is always essive. This means that the topic should be interpreted as a noun phrase that starts with 'there is', 'there are'. E.g.:

[ ha doe]
 * hA dOE
 * (TOPfire)
 * => to burn / there is (a) fire.

Patient

 * → Corresponding morpheme: iA [ia].

The patient is the syntactical object that is affected by the topic. It has no active participation in the process expressed by the topic. It will generally translate to an (in)direct object or prepositional object whenever the topic expresses an action, or the subject if the topic describes a (passive) state. E.g.:

[ ia 'anoæsi hadoe]
 * iA aNoWsIhAdOE
 * (PAThouse TOPfire)
 * "fire is happening to the house"
 * => the house burns / the house is on fire.

Agent

 * → Corresponding morpheme: lA [la].

The agent is the syntactical object that initiates the topic. The appearance of an agent usually turns sentences into (transitive) actions. E.g.: [ la æma ia'anoæsi hadoe]
 * lA 1mAiAaNoWsIhAdOE
 * (AGme PAThouse TOPfire)
 * "I initiate the fire that is happening to the house"
 * => I set fire to the house.

Modifier

 * → Corresponding morpheme: iR [ir].

A modifier adds information to a sentence which cannot be covered by one of the previous functions. It is safe to say that a modifier corresponds to an adverbial constituent: time, place, manner, cause, etc. It is important to know that modifiers harbour the key to expressing time tenses, since this cannot be done through verb conjugation. E.g.:

[ ir asdale laæma ia'anoæsi hadoe]
 * iR aSdAlElA1mAiAaNoWsIhAdOE
 * (MODfuture AGme PAThouse TOPfire)
 * "In the future I initiate the fire that is happening to the house"
 * => I will set fire to the house.

Subordinate functions
Subordinate functions can only appear in combination with one of the primary functions. They are therefore always tied to one specific argument of the sentence. There are 2 subordinate functions.

Characteristic

 * → Corresponding morpheme: eM [em].

A characteristic provides additional information of a specific syntactical argument. It is in some way comparable to a modifier. But, as mentioned before, a characteristic is related to a specific argument in the sentence, not the sentence as a whole. A characteristic will generally translate to an adjective component, which can contain information about size, shape, age, a visual feature, or much more. E.g.: [irasdale laæma ia'anoæsi iaem vera hadoe]
 * iRaSdAlElA1mAiAaNoWsI iAeM vErAhAdOE
 * (MODfuture AGme PAThouse PAT-CHARpreciousness TOPfire)
 * "In the future I initiate the fire that is happening to the house which is precious"
 * => I will set fire to the precious house.

Referent

 * → Corresponding morpheme: vW [væ].

The referent is a syntactical element that places an argument in a semantical relation to something else. The referent is possibly the least plausible function for Western learners, as it corresponds to seemingly irrelevant structures. Therefore, it is easiest to analytically translate it as "when compared to X". The referent can express a comparative or superlative structure (e.g.: I am big, when compared to you --> "I am bigger than you"); it can express a relative indication in time or space (e.g.: in the front, when compared to me --> "in front of me"); but it can also simply express a possessive pronoun (when compared to me, the father --> "my father"). E.g.:

[irasdale laæma ia'anoæsi iaemvera iavæ ema hadoe]
 * iRaSdAlElA1mAiAaNoWsIiAeMvErA iAvW 2mAhAdOE
 * (MODfuture AGme PAThouse PAT-CHARpreciousness PAT-REFyou TOPfire)
 * "In the future I initiate the fire that is happening to, when compared to you, the house, which is precious"
 * => I will set fire to that precious house of yours.

Free functions
The free functions are functions aren't a part of the hieratic relationship between the primary and subordinate functions. They are free (hence the name). There are 2 free functions.

Sentence bracket

 * → Corresponding morphemes: lW [læ] + iW [iæ].

The sentence bracket is a set of two morphemes that allow subordinate clauses to be constructed within the main clause. The opening bracket 'læ' is comparable to the relative pronouns 'that' and 'which', and the closing bracket 'iæ' is used to mark the end of the subclause. The entire sentence bracket will be subordinate to the root word to which it is suffixed, and, because otherwise it would be superfluous, the sentence bracket will in turn contain primary and/or subordinate functions itself. E.g.: [irasdale laæma ia'anoæsi iaemvera iavæ'ema iaem læ la'ema haeleanara iæ hadoe]
 * iRaSdAlElA1mAiAaNoWsIiAeMvErAiAvW2mAiAem lW lA2mAhAeLeAnArA iW hAdOE

In this sentence, the subordinate clause translates to you love. The bracket links to a characteristic, which in turn is linked to the patient of the main clause (the house). Hence the formulation iAeMlW [ia-em-læ].


 * (MODfuture AGme PAThouse PAT-CHARpreciousness PAT-CHAR- BRACKET.OPEN(AGyou TOPmuch-love BRACKET.CLOSE) TOPfire)
 * "In the future I initiate the fire that is happening to, when compared to you, the house that you love, which is precious"
 * ==> I will set fire to the precious house of yours that you care for so deeply.

Sentence breaker

 * → Corresponding morpheme: tA [ta].

The final function is the sentence breaker. Is is a fairly important function that covers for the absence of interpunction, as well as the language's asyndetic tendency (which is a fancy way of saying that there is no word for the conjunction and). The sentence breaker morpheme can be placed between to sentences with the same effect as a comma or full stop in most other languages, but it can also appear within sentences to create the following effect: [laæma haen laSem haer]
 * lA1mAhAeNlAksEMKhAeR
 * (AG me TOPvision AGSam TOPhearing)
 * => Sam and I are looking and listening.

[laæma haen  ta laSem haer]
 * lA1mAhAeN tA lAksEMKhAeR
 * (AG me TOPvision (sent.br.) AGSam TOPhearing)
 * => I am looking and Sam is listening.

Conciseness
The Ælis pattern for function marking tends to create very large sentences, which is why discourse aims to be as concise as possible. Conciseness is achieved on two levels.

Omission of functions
SVO languages usually do not allow the subject or verb to be omitted, but since Ælis' active-stative morphosyntactic alignment in combination with a thorough topicalisation makes omission possible for all functions without losing grammatical correctness, discourse aims to omit as many functions as possible. If in a conversation the same things are discussed over a period of several sentences, speakers may assume that omitted words/functions remain the same as in the previous sentence(s) unless otherwise specified. Elliptical omission of this type can strongly cut the length and weight of sentences. A clarifying example:


 * Person A: iRaSdAiOlA1mAiAaNoWsIiAvW2tE
 * [irasdaio laæma ianoæsi iavæ'ete]
 * (MODpast AGme PAThouse PAT-REFyou)
 * This sentence doesn't contain a topic, so it doesn't explicitly express an action.
 * ==> "I (did something) to your house."


 * Person B: hAnE
 * [hane]
 * (TOPquestion)
 * Person B asks A to specify a topic.
 * ==> "What (did you do to it)?"


 * Person A: hAdOE
 * [hadoe]
 * (TOPfire)
 * Person A specifies a topic.
 * ==> "I set fire to it."


 * Person B: iRaRnE
 * [irarne]
 * (MODreason-question)
 * Person B asks A to specify a modifier of causality.
 * ==> "Why (did you do that)?"


 * Person A: iRaReleAnAsI
 * [irareleanasi]
 * (MODreason-hate)
 * Person A specifies a modifier of causality. He doesn't specify who hates whom, but the context provides a plausible option.
 * ==> "(I did it) because I hate you."

Note that A's first reply ([hadoe]) is the exact same formulation as the very first example sentence of this page, which meant "There is a fire". Only now, the implied meaning is completely altered by the context.

If a speaker wishes to start talking about something completely different, and the other speaker has a reason to falsely suppose a coherence between the two sentences, in spite of the eventual use of a sentence breaker [ta], they can use the formula iRrWuA [irræua], which would literally mean "irrelevantly", or something along the lines of by the way or on a side-related note.

Omission of function markers
There are some cases where the function markers themselves may be omitted completely. This is the case with short, generic formulations such as thanks or you're welcome, as such utterances appear frequent and are generally unambiguous. For example: while the sentence hAgIrAlA1tEiA2tE [hagira laæte iaete] (for you there is total gratitude on my behalf) would be the formal way of thanking someone, colloquial speech will generally abbreviate this to simply gIlE [gile] (a lot of gratitude) or gIrA [gira] (total gratitude). Similar expressions without function markers include:
 * hInArA [hinara] or nArA [nara] (Hello/Goodbye)
 * nWnE [næne] (What's up?)
 * dIrA [dira] (You're welcome)
 * iSnE [isne] (What's next?)
 * rEAnE [reane] (What's (your) name?)
 * kmERIK [rea'meri] ((My name is) Mary)

Flexibility of function attribution
In many situations, function markers can be attributed in a variety of ways while still expressing the same thing. It is misleading, for instance, to think that a possessive pronoun in an English sentence automatically corresponds to a referent in Ælis. Although it would be correct to say hArEAKhAvW1nIhAeMkaNAK [hareahavæ'ænihaem'ana'] (lit.: There is a name, which, when compared to me(♀), is Anna --> "My name is Anna"; it is preferable to say hArEAkaNAKiA1nI [harea'ana'ia'æni] (lit.: The being named Anna is for me(♀) --> "My name is Anna"; since this utterance contains 30% less root words than the first.

Root word list and vocabulary types >