Distributed morphology
The Distribuierte morphology ( engl. Distributed Morphology, abbreviated DM, also: distributed morphology) is a theoretical concept of general linguistics, in which the morphology is distributed on the syntax and phonology. With their different flexion but also derivational phenomena to be explained in the languages of the world.
- 3.1 The structure of the lexicon
- 3.2 Elsewhere
- 3.3 Readjustment Rules
- 3.4 Merger, demerger, merger, poverty 3.4.1 merger
- 3.4.2 cleavage
- 3.4.3 merger
- 3.4.4 depletion
Introduction
The Distribuierte morphology includes a conceptual framework of different models of the processes that occur at the interface between syntax and phonology. These models attempt to explain how abstract morphosyntactic and morphosemantische features are replaced by their concrete phonetic realization. The model in its original form dates back to two articles of the American linguist Morris Halle and Alec Marantz from the years 1993 and 1994.
The Distribuierte morphology is the taxonomy of Greg Stump (2001 ) to a post- syntactic realisational - lexical theory of morphology.
Post- syntactic means that the selection of the interpreted in the establishment morpho - syntactic features based on the hierarchies of syntactic derivation. This means that the syntactic part of the morphology may all access the parent structure of the currently processed node terminal.
Realisational means that the morpho - syntactic features are independent entities of the structure. In the morphology of these features can be realized, by replacing them by the phonological features. These are unlike the incremental theories, in which the morphology introduces the features only in the structure.
Lexical means that the information that morphosyntactic features are replaced by which phonological, morphemweise are stored as separate entities in the mental lexicon. Thus, for example, would the association [ plural ↔ / -s / ] an entry in the mental lexicon of an English- speaker. These stored in the lexicon entries, which associate morphosyntactic phonological features with chains, are in the DM as vocabulary items (such as: vocabulary entries, in another VI) respectively. In inferential theories, however, the root of a word is associated with specific rules with a separately stored in the lexicon word form that is sensitive to certain characteristics.
The course of a syntactic derivation according to the model of the DM proceeds as follows: First, a syntactic structure on abstract morphosyntactic categories and features is generated. After completion of the syntactic derivation, the terminal symbols of the resulting structure by chains phonological symbols (ie sounds or the representations of sounds ) replaced. This linking of syntactic categories and their phonological correspondences is commonly referred to as morphology. In the DM morphology itself is not considered as an independent component of the language, but as a process that is distributed over the levels of syntax and phonology, hence the name distribuierte ( = distributed) morphology. Goal of DM is to explain the principles by which this substitution is going on.
The Distribuierte morphology is to be understood less as an independent theory, but rather as a class of approaches that share a set of concepts. The three basic building blocks, drawing on the DM of each derivative are Late establishment (English late insertion), sub- specification (English underspecification ) and syntactic structure all the way down (English syntactic structure all the way down). These are explained in the following section.
Basic Concepts
The following three concepts are considered to be basically lying for Distribuierte morphology.
Late establishment
The principle of the late establishment (English late insertion) states that the replacement of morphosyntactic properties and features by their phonological correspondences occurs late in the derivation. In theory, the setting takes place after completion of the syntactic structure construction. In the T- model of generative grammar theory of the date of establishment is located at the beginning of the PF- path, so the morphology is in the DM model viewed as an interface phenomenon between syntax and phonology.
Under specification
The principle of sub- specification results from a number of assumptions which specifically are not even for the DM and some are as old as the modern theoretical linguistics itself only in its combination that they form an essential basic ingredient of each DM -like theory. These assumptions include the subset principle, the principle of Merkmalsdekomposition and the concept of specificity.
The subset principle
The establishment, ie the replacement of the morphosyntactic features and morphosemantic by their phonological correspondences is done, after the derivation of syntactic structure taking into account the so-called subset principle. This principle states that a VI is inserted into a terminal node, as soon as the amount of features on this node are a superset of the features that are specified in the VI. It is said that the characteristics are specified on the VI.
Carries a terminal node such as the characteristics [ 1 Person, singular, active, present tense, indicative mood ], as a marker / e /, which for the characteristics [ 1 Person, singular, active, indicative ] is specified, are used in this terminal node, as its feature set is a subset of the feature set on the terminal nodes.
An effect of the subset principle is that syncretism can be derived systematically with him. Syncretism are occurrences of the same marker in different morphosyntactic contexts. Thus, the above marker / e / both in the indicative, and in subjunctive contexts emerge, since it satisfies the subset principle in both contexts.
Merkmalsdekomposition
Characteristics are assembled in theory usually from smaller units. In German, as in all Germanic languages , there is a syncretism between the verb forms of the first and the third person plural (for example, we / they go -en ). As with privative features this syncretism would not dissipate, it is believed that these features have something in common. The personal characteristics mentioned for example, have in common is that they are not the einschlissen or the addressee of a speech act ( second person). These people features may therefore be composed as follows:
- First Person: [ 1, -2 ]
- Second person: [-1, 2 ]
- Third person: [-1, -2 ]
This splitting of features into smaller units called Merkmalsdekomposition. The marker / - s /, which is appended to verbs in the first and third person plural in the Germans, can be specified by this characteristic system for the characteristics [-2, plural ]. After the subset principle it can be used both in first plural - as well as in third - plural contexts. This type of feature composition can be apply to almost all conventional features, although the nature of the underlying characteristics from batch to batch is different.
Specificity
By Merkmalsdekomposition and subset principle, it is possible that several VI can be used in einunddenselben context. For example, if a VI for [ 1, -pl ] specifies another for [ 1 ], so both could be used in a node with the features [ 1, - pl, indicative ]. In order to control such cases one takes the principle of specificity at that predicts which of the two markers will be used. How this looks depends on the chosen approach. In the original version of the DM of the markers was used, which is specified for most of the features. In above example, the first marker would be used, as it is specified for two features, while the second, it is only for a characteristic. The system reaches its limits when a third VI is added, which is specified, for example, [ 1, indicative ]. In such cases, additional assumptions regarding the specificity are needed. Most approaches to use this opportunity characteristic hierarchies into play. Thus, the marker is used, the characteristics of which are higher in the hierarchy than the other marker. As these hierarchies look also depends greatly on the selected approach, but usually these hierarchies are based on typological empirical data.
Syntactic structure all the way down
Another principle behind DM is as Syntactic Hierarchical Structure All the Way Down (English, mutatis mutandis: Syntactic hierarchical structure all the way down ) respectively. It essentially says that the morphological and syntactic components of a language based on the same hierarchical structures. The elements with which both morphology and syntax working are to be understood as a discrete constituent, rather than results of morphological processes.
Further concepts and background assumptions
The structure of the lexicon
The hitherto unknown concept of a mental lexicon is rejected by the authors of the distributed everywhere morphology. Phenomena that have been explained so far as lexically, such as the formation of composites or derivatives are dispersed in the DM to other components. In the mental lexicon itself the pure associations of morphosyntactic interpretable units and their meanings are only stored. In order to avoid overlaps with other concepts of the mental lexicon, one therefore speaks in the DM of an " encyclopedia " in which these associations are stored. The information stored in them are tripartite: First (VI german vocabulary item, ) is in a semantic- conceptual meaning with a so-called entry associated vocabulary. The VI are again pairs, which associate morphosyntactic and / or morphosemantische features with their respective phonological representation. In the literature, these VIs have mostly the form of context-sensitive setting rules:
As an example, the VI for the suffix is specified, which in the first person singular present active marks ( I go -e) in German verbs:
The notation indicates that the feature combination [ 1 Person, singular, active, present tense ] is sound / e / expressed by the and that this sound is a suffix (denoted by the symbol "-" in front of the sound). The Encyclopedia entry for this morpheme thus contains the information that the morpheme is (actually as a schwa [ ə ] ) realized as / e / and the characteristics of first person, singular, active, present tense expresses is associated with the meaning that the speaker alone (1st singular) of talk time ( present tense) the performer (Active) is an action ( verb ending ).
Furthermore there are two types of lexical entries for a l- morphemes and on the other the f- morphemes. While the functional f - morphemes are fixed, the l - morphemes serve as a placeholder for any lexical content. Thus, strains of nouns, verbs or adjectives (each to a specific language -related) l - morphemes. These can be exchanged at will, without changing the basic grammaticality of an expression. f - morphemes, however, are firmly connected to the grammatical construction, that is f - morphemes can not be easily replaced if you do not change the underlying features. To illustrate, the following example is intended to serve:
In this set of three l - morphemes are included, namely, see Mary as the root of the subject, see, as the root of the verb, and man as the root of the object. These three can be ( in certain dimensions ) replaced by any other, without changing the grammaticality of the sentence:
The other morphemes of set up, -t and the other hand, are firmly connected to the syntactic structure, changes to these is the sentence ungrammatical:
This is to illustrate that, in contrast to the l - morphemes, which are bonded to the concrete morphosyntactic context f - morphemes. The f - morphemes include not only the Flexionsaffixen all kinds of pronouns, Determinierern (that is, for example, definite and indefinite articles), adpositions and other functional categories.
The contextual, almost unlimited versatility of the l - morphemes allows you to use one and the same root in different categories, ie, an l - morpheme as "les", depending on the context into a noun ( the reading), a verb ( we read ) a participle ( the reading man) or the like. As what is used a root is determined by the morpho-syntactic context, but does not go from the root itself. If the nearest neighbor of a terminal node in which the root is to be used, for example, a tense node, the root is interpreted as a verb stem, the nearest neighbor is a Determinativknoten, the tribe as a noun is, he is a nominal node as an adjective interpreted etc.
Elsewhere
In many variants of the DM the existence of a so-called Elsewhere marker is assumed. This is a VI, which fits on all presently contemplated contexts and is used wherever no other, higher specified VI fits. The possibility of the existence of a Elsewhere marker results from the other assumptions, so the subset principle is often referred to as the Elsewhere Principle.
Readjustment Rules
The term " readjustment rules" refers to rules that once on the VIs used to operate after the actual establishment to subject them phonological changes. These rules are applied to morphologically related Allophonie. This is the phenomenon that a particular phonological rule applied only in the presence of a specific morphological marker, ie in a not purely phonological context.
To illustrate an example from the German verb inflection is intended to serve.
The affix on a verb in the second person plural is / - t / (her go - t), which is the second person singular /-st / (du geh- st).
With certain verbs, however, a schwa ( [ ə ], written as
Merger, demerger, merger, poverty
In the literature a number of other DM rules are adopted, which manipulate the syntactic structure before the VIs are used.
Merger
The merger operation moves a terminal node to a different position in the hierarchy. This function is assumed to linearization phenomena to explain. Such occurs when the architecture of the theory, the licensing of a feature to a different position than at the providing at which the feature is implemented. In some models of syntax, it is believed that certain terminal nodes can only get the value of a feature when they are at a certain position within the syntactic structure. Now occurs in the case that this assumed position is different from that one perceives when hearing the finished set, one falls back on merger operations and shifts the respective terminal node to the empirically observed position.
Cleavage
Cleavage (English fission ) the splitting of a terminal node is freely understood according to the principle of the specification. Carries a terminal node, the characteristics [a, b] so you can, in the absence of a specified characteristics for both VIs two VIs are used: one that is specified for [ a], and another, which is specified for [b ].
For example, while Halle and Marantz still considered as a separate division, controlled by a control operation, other authors derived from a concept of division of the subset principle. Thus, features that are not even realized the most highly specified and appropriate VI to be used to set up less specific VIs.
Merger
Under merger ( engl. fusion ) is understood in the DM an operation, will be merged with the two terminal nodes to a single one. In this way, for example, explains portmanteaus, which are single morphemes, which realize the characteristics of several separate se terminal nodes simultaneously.
Impoverishment
As depletion (English impoverishment ), the deletion of individual features from the structure understood. Most of these rules are context -sensitive. In this way, a less specific VI, in a context for which there is a highly specified VI, are used as the context for the higher specified VI is destroyed by the depletion. Since less specific VIs fit more contexts, it is called on the application of impoverishment rules from so-called Retreat to the General Case (engl, about returning to the general case).
Jochen Trommer considered impoverishment as establishment of highly specific marker zero. He considered poverty as a consequence of the principle of division.
The Distribuierte morphology and other models of grammar
Nevertheless, the DM of generative terminology, such as the schema of the T- model, and some other assumptions used this, the two models are only partially compatible. One of the main differences is that all the features are available during the syntactic derivation, including those that are interpreted only by the morphology itself, such as characteristics of inflection classes. In the theories according to Noam Chomsky, however, the syntax is free of any features that may not even be from the syntax interpreted, which include Flexionsklassenmerkmale. Another difference concerns the timing of the morphological interpretation. While Chomsky prefers a pre- syntactic morphology, finds the morphological interpretation in the DM instead of post- syntactically.
Confines
Heidi Harley and Rolf Noyer, construct in their introductory essay on DM 1999 Language " Martian ", which can not explain the DM. In this fictional language the feature [ plural ] is expressed by the last syllable of the word - regardless of whether this syllable to a Flexionsaffix, the tribe or to a combination of several of these belongs - is repaid. In the DM such " Readjustment Rules" can affizieren only individual vocabulary entries, but not more than one at a time. The consequence of this is that a language that expresses morphological features in this way, can not exist.
Which occurs in the speech pattern, which is referred to as subtraction, as well as the truncation corresponds to a large extent to this pattern, but always relates to the trunk, and never affixes. An example of a language with such a phenomenon is the Uto- Aztec language Papago, in which the perfective is marked by the paying off the last syllable of the stem.
Even more cases of non- concatenative morphology often can not be derived without additional assumptions in a DM approach.
Also problematic is the melding of cases extended Exponenz with certain modifications of the DM approach. Under Extended Exponenz means the realization einunddesselben feature by more than one marker. If you, like most DM approaches that VIs the morpho - syntactic features " eat up " at the institution, so that they are no longer available for subsequent setting available, the occurrence of extended Exponenz should be excluded. Also on this problem there are a number of approaches, such as context-sensitive setting rules, secondary Exponenz or - in analogy to the impoverishment rules - enrichment rules that copy the multiple feature realized in certain contexts.
One problem with the adoption of a radically underspecified Elsewhere marker in combination with the subset principle -based splitting concept is that the Elsewhere markers could potentially be used as often in einunddenselben terminal nodes. To prevent this, a filter for such cases it must be assumed that any of the following principles of the DM, and thus represents a potential weak point in the theory.
The strict separation of the two types of morphemes ( l -and f- morphemes ) leads to the problem that some features may not be available until after the setting of l - morphemes, but this sometimes in the syntax already need to be considered, such as genus characteristics. These are, for example, in German, passed down within a nominal projection, that is, in the entire noun phrase are Genus features available, which can be seen at the Genuskongruenz of adjectives and articles with the nouns. What genus is realized, however, in the end, decided until after the establishment of the nominal strain, also here are a number of non-trivial additional assumptions needed to derive the correct derivation of such a construction.