Janez Orešnik CDU 811.111'366.543 
University of Ljubljana CDU 811.111'367.626 

NATURALNESS IN ENGLISH: 
(A) TIIE GENITIVE, (B) THE PRONOUNS 
In Slovenia, the natura[ syntax ofthe Klagenfurt brand has been extended to the study ofthe behav­iour of (near-)synonymous syntactic expressions, here called syntactic variants. The work below is illustrated by the (morpho)syntax of the English genitive and the English pronouns. The language material is divided into consecutively numbered deductions in each ofwhich the existence ofa (mor­pho)syntactic state of affairs is predicted on the basis of apposite assumptions and Andersen's markedness alignment rules. The basic point: given two (morpho)syntactic variants, such that one of them shows feature A, and the other shows feature B, the theory can answer the question as to which ofthe two variants shows which ofthe two features A and B. 
Introduction 
The subject-matter of my paper is a (language-universal) theory developed in Slovenia by a small group oflinguists (under my guidance), who mainly use English, German, and Slovenian language material as the base ofverification. Our work owes much to, and exploits, the (linguistic) Naturalness Theory especially as elaborated at some Austrian and German universities; cf. Mayerthaler (1981), Wurzel (1984), Dressler et al. (1987) and Dressler (2000). Naturalness Theory has also been applied to syntax, notably at the University of Klagenfurt; the basic references are Dotter (1990), Mayerthaler & Fliedl (1993) and Mayerthaler et al. (1993; 1995; 1998). Within the natural syntax of the Klagenfurt brand, the Slovenian work group has constructed an extension that studies the behaviour of (near-)synonymous syntactic expressions, here called syntactic variants. Whenever two syntactic variants are included in the same naturalness scale, and consequently one variant can be assert­ed to be more natural than the other, something can be said about some grammati­cal properties of the two variants. 
Within Naturalness Theory, Mayerthaler (1981:10 et passim) distinguishes sem­and sym-naturalness. Because the present paper utilizes sem-naturalness only, Mayerthaler's distinction will not be discussed. Sem-naturalness will simply be called naturalness in the continuation of the paper. The predicate "natural" will be defined as simple (for the speaker) from the cognitive point ofview. This kind ofnat­uralness is similar to traditional markedness, and the following approximate equa­tion can be stated as a first orientation of the reader: cxmarkedness = -anaturalness. It is practically impossible to compare markedness and naturalness in (morpho)syn­tax seeing that the application of both in that field is in a sta te of flux. 
Naturalness values will be stated in naturalness scales. The basic format is >nat (A, B)-i.e., with respect to cognitive complexity, A is more natural than B. This is the speaker's viewpoint. It is further assumed that, from the hearer's viewpoint, B is more natura! than A. (This is based on the assumption that the interests of the speaker and the hearer in a communicative situation are antagonistic.) Consequent­ly, the scale >nat (A, B) can be substantiated either by showing that A is more natu­ra! than B for the speaker, or by showing that B is more natural than A for the hear­er. However, the matter plays a minor role in this paper, and will not be elaborated. 
To cover any optional usage ofA or B in >nat (A, B), this framework assumes the following two additional formats derived from the basic format: 
(i) 	
>nat (A + B, B)-i.e., with respect to cognitive complexity, the optional use of A (with respect to B) is more natura! than the use of B on its own; 

(ii) 	
>nat (A, A + B)-i.e., with respect to cognitive complexity, the use ofA on its own is more natura! than the optional use of B (with respect to A). 


Any scale in one of the two derived formats (i-ii) is asserted to be true whenever the corresponding scale in the basic format >nat (A, B) is asserted to be true. Therefore, when a scale couched in a derived format is used, it suffices to back up the corresponding scale in the basic format. Given the wealth of optional usage in languages, the applicability of my framework would be greatly reduced without the two additional formats. 
In the present paper, the language examples are dealt with in "deductions". Each deduction contains at least two naturalness scales. The naturalness values of paired scales will be aligned by the principle of markedness alignment as stated in Andersen 1968 (repeated in Andersen 2001), and adapted to naturalness in the fol­lowing way: what is more natural tends to align with another instance of more nat­ura!, and what is less natura! tends to align with another instance of less natura!. 
The theory utilizes the following ways of determining naturalness in (morpho} syntax: 
(a) 	
The principle of least effort (Havers 1931:171). What conforms better to this principle is more natura!. What is cognitively simple (for the speaker) is easy to produce, easy to retrieve from memory, etc. 

(b) 
Phylogenetic age. What is older phylogenetically is more natura!. What is cogni­tively simpler (for the speaker) is acquired earlier by the language. 

(c) 	
Prototypicality. What is nearer to the prototype is more natura!. 

(d) 
Degree of integration into the clause. What is better integrated into its clause is more natura!. This partially exploits (c): the prototypical syntactic situation is for a syntactic element to be well integrated into its syntactic construction. 

(e) 	
Frequency (in the spirit of Fenk-Oczlon 1991). What is more frequent token­and/or typewise is more natura!. What is cognitively simpler (for the speaker) is used more. 

(f) 	
Small v. large class. The use of a unit pertaining to a small class is more natural than the use of a unit pertaining to a large class. During speech smaU classes are easier for the speaker to choose from than are large classes. 

(g) 	
Specialised v. non-specialised use. The specialised use of a category is more nat­ural than its non-specialised use. This generalisation is based on the following consideration. All kinds of categories occur in the most natural lexical items, paradigms and constructions of the language, and ebb on the way out of that core. Take for example a language whose noun phrases distinguish singular, plu­ral and dual. Although singular, plural and dual are not equally natural with respect to one another, each of them is highly natural in its own field. Por instance, the dual is highly natural (specialised) as an expression of duality: >nat (dual, singular/plural) /in expressions of duality. This is correlated with the cir­cumstance that all three numbers are present in personal pronouns, i.e. in the most natural noun phrases, while they may be present to different degrees in the remaining noun phrases of the language. (Recall the above-mentioned align­ment rules.) Por the relevant typological <lata about the grammatical numbers, see Corbett (2000). 

(h) 	
Use v. non-use. The use of a category is more natural than its non-use. With this principle it is possible to fix the cutoff point between the use and non-use of a category. Because the use of a category normally ( also) occurs with the most nat­ura! units of the relevant kind, the rules of alignment force the assumption that the use of a category is more natural than its non-use. E.g. >nat ( +dual, -dual) / in expressions of duality. See the preceding item (g). 

(i) 	
Acceptable v. non-acceptable use. What is acceptable is more natural than what is not acceptable. The very reason for the acceptability of a syntactic unit is its greater naturalness with respect to any corresponding non-acceptable unit. 


G) 	What is more widespread in the languages of the world is more natural (the typological criterion). What is cognitively simpler (for the speaker) is realized in more languages. 
At present tirne, the above items (a-j) are the only ways used by the theory to de termine naturalness in (morpho )syntax. In this sense the theory is highly con­strained. Any recourse to additional criteria should be viewed as a weakness of the theory. (The current version of the items (a-j) has resulted from cooperation with Helena Majcenovič.) 
Illustrations of some of the above items (a-j) will be adduced as this article pro­ceeds. 
The framework just outlined will now be applied to some (morpho )syntactic vari­ants of English. Pairs of variants have been determined on the basis of my linguistic experience. The upper limit on the length of a variant is two linked clauses. As already mentioned, each case considered is presented in the format of a deduction. (The ordering of the deductions is mostly arbitrary.) 

Examples 
The examples (these are meant to be simple and variegated) deal wiLth various aspects of the genitive and of the pronouns. 
(A) 
The Genitive 

l. 
English. The genitive case shows a desinence in the singular only. (See deduction 


2 for a qualification.) E.g. cat's v. cats' (Biber et al. 1999:292). The two syntactic variants: the singular and the plural of the genitive case. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat ( +, -) / marked genitive case 
I.e. marked ( = with marking) genitive case is more natural than unmarked ( = without marking) genitive case. The use of a category is more natura! than its non­use. The use of marking for case is more natura! than its non-use. See item (h) in the Introduction. 
1.2. >nat (singular, plural) 
I.e. the singular is more natural than the plural (Mayerthaler 1981:15). By the principle ofleast effort; see item (a) in the Introduction. Whereas the plural is often encoded by special means, the singular is often left bare (in many languages). 
The >nat (= high naturalness value) of scale 1.1 is the marked (= with marking) genitive case. It is aligned with the >nat ofscale 1.2, which is singular. The <nat (= low naturalness value) of scale 1.1 is the unmarked (= without marking) genitive case. It is aligned with the <nat of scale 1.2, which 1s plural. 
3. The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference between the singular and the plural ofthe genitive case, 
such that one grammatical number marks the genitive, and the other grammatical number does not, it is the singular that tends to mark the genitive, and it is the plu­ral that tends not to mark the genitive. Q.E.D. 
As can be seen from the above deduction, my theoretical framework does not contain any generative component and operates ex post facto. 1 cannot predict the existence of the genitive with a desinence (cat's) and of the genitive without a desinence (cats}; 1 cannot predict that one kind of genitive is used in tlie singular, and the other kind of genitive is used in the plural. However, if this data is given, 1 can predict that it is in the singular that the genitive showing a desinence is used, and that it is in the plural that the genitive lacking a desinence is used. The reverse situation is not likely to obtain. It is not probable that the genitive lacking a desinence would be used in the singular, and the genitive showing a desinence would be used in the plural (ifthe desinence is the differentia specifica ofthe two genitives). It is such predictions (that is, synchronic accountsj"explanations") that constitute the chief motive of my work. Mutatis mutandis, these remarks apply to all deductions of the present paper. 
The basic point can be stated as follows. Given two (morpho)syntactic variants, such that one of them shows feature A, and the other shows feature B, the theory can answer the question as to which of the two variants shows which of the two fea­tures A and B. 
2. English. In the plural, the genitive case equals the common case in phonological form, e.g. cats' is pronounced as cats. Only irregular plural nouns take a desinence, 
e.g. 
men's (Biber et al. 1999:292). The two syntactic variants: the additive and the interna! plural formations. 

l. 
The assumptions ofNaturalness Theory: 




1.1. >nat ( +, -) / marked genitive case 
Le. the marked (= with marking) genitive case is more natura! than the unmarked (= without marking) genitive case. The use of a category is more natura! than its non­use. The use of marking for case is more natura! than its non-use. See item (h) in the Introduction. 
1.2. >nat (interna!, additive) / formation of plural 
Le. interna! formation of the plural is more natura! than additive formation of the plural (Mayerthaler 1981:25). The additive formation of the plural is easier for the hearer lto process. The situation is the reverse for the speaker, who retrieves ready interna! formations of the plural from memory. This is a natura! situation by the principle of least effort; see item (a) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 


2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference between additive and interna! plural-noun formations, 


such that one formation marks the genitive, and the other formation does not, it is the interna! plural formation that tends to mark the genitive, and it is the additive plural formation that tends not to mark the genitive. Q.E.D. 
3. English. Specifying genitives, e.g. the girl's face, are prevailingly definite. Classifying genitives, e.g. a bird's nest, are prevailingly indefinite (Biber et al. 1999:294-5). 
The two syntactic variants: specifying and classifying genitives. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (specifying, classifying) / genitive 
Le. a specifying genitive is more natural than a classifying genitive. The specify­ing function is the most important function of the genitive (Biber et al. 1999:294). 
Thus the specifying function is very likely the prototypical function of the genitive. See item ( c) of the Introduction. 
1.2. >nat (+, -) / definite NP 
Le. a definite noun phrase is more natural than an indefinite noun phrase. A def­inite noun phrase is more accessible for the speaker than an indefinite noun phrase. By the principle of least effort; see item (a) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 

2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference between specifying and classifying genitives, such that 


one kind of genitive is prevailingly definite, and the other indefinite, it is the speci­fying genitives that tend to be definite, and it is the classifying genitives that tend to be indefinite. Q.E.D. 
4. English. The classifying genitives commonly occur in the plural, e.g. boys' camp 
(Biber et al. 1999:295). The two syntactic variants: the specifying and the classifying genitives. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (specifying, classifying) /genitive 
Le. a specifying genitive is more natural than a classifying genitive. The specify­ing function is the most important function of the genitive (Biber et al. 1999:294). Thus the specifying function is very likely the prototypical function of the genitive. See item (c) of the Introduction. 
1.2. >nat (singular, plural) 
Le. the singular is more natural than the plural (Mayerthaler 1981:15). By the principle ofleast effort; see item (a) in the Introduction; whereas the plural is often encoded by special means, the singular is often left bare (in many languages). 
A special case of 1.2: 
1.2.1. >nat (singular & plural, only plural) 
Le. admitting both the singular and the plural is more natural than admitting only the plural. The scale has the format >nat (A + B, B); see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 

2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1, 1.2.1 and 2 it can be deduced: If there is any difference between the specifying and the classifying genitives, 


such that one kind of genitive is used both in the singular and in the plural, and the other kind of genitive is used only in the plural, it is the specifying genitives that tend to be used both in the singular and in the plural, and it is the classifying genitives that tend to be used only in the plural. Q.E.D. 
5. English. Whereas specifying genitives can be paraphrased by an ofphrase, e.g. the girl'sjace -thejace ojthe girl, classifying genitives are frequently paraphrased by ajor­phrase, e.g. children's clothes -clothesjor children (Biber et al. 1999:294-5). 
The two syntactic variants: the specifying and the classifying genitives. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (specifying, classifying) / genitive 
Le. a specifying genitive is more natural than a classifying genitive. The specify­ing function is the most important function of the genitive (Biber et al. 1999:294). Thus the specifying function is very likely the prototypical function of the genitive. See item (c) ofthe Introduction. 
1.2. >nat (the type thejace ojthe girl, the type clothesjor children) 
Le. the type the jace oj the girl is more natura! than the type clothes jor children. For is a more specific preposition than oj, therefore more satisfactory than ojfor the hearer. The situation is the reverse for the speaker; see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 



2.1. >nat tends to aligµ with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: If there is any difference between specifying and classifying genitives, such that 


one kind of genitive can be paraphrased by an ofphrase, and the other kind by ajor­phrase, it is the specifying genitives that tend to be paraphrased by an ofphrase, and it is the classifying genitives that tend to be paraphrased by ajor-phrase. Q.E.D. 
6. English. Elliptic genitives -such as in That isn't my handwriting. It's Selina's! -are 
mostly specifying (Biber et al. 1999:296-7). The two syntactic variants: specifying and classifying genitives. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (specifying, classifying) / genitive 
Le. a specifying genitive is more natura! than a classifying genitive. The specify­ing function is the most important function of the genitive (Biber et al. 1999:294). Thus the specifying function is very likely the prototypical function of the genitive. See item (c) of the Introduction. 
1.2. >nat ( +, -) / ellipted head noun of genitive Le. an ellipted head noun of the genitive is more natura! than a non-ellipted head 
noun ofthe genitive. By the principle ofleast effort; see item (a) in the Introduction. A special case of 1.2: 
1.2.1. >nat (+/-, -) / ellipted head noun of genitive 
Le. a genitive admitting ellipted and non-ellipted head nouns is more natural than a genitive admitting only non-ellipted head nouns. The scale has the format >nat (A + B, B); see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 

2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1, 1.2.1 and 2 it can be deduced: Ifthere is any difference between specifying and classifying genitives, such that 


one kind of genitive associates with ellipted and non-ellipted head nouns, and the other kind associates with non-ellipted head nouns, it is the specifying genitives that tend to associate with ellipted and non-ellipted head nouns, and it is the classifying genitives that tend to associate with non-ellipted head nouns. Q.E.D. 
7. English. Many independent genitives have become conventionalized, so that they need no supporting head noun in the context, e.g. she's going to a friend's. Indepen­dent genitives are found particularly in conversation (Biber et al. 1999:297, 300). The present deduction is continued in deduction 8. 
The two variants: conversation and the written registers. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (conversation, written registers) 
Le. conversation is more natural than the written registers (Dotter 1990:228). Oral communication is the primary form of communication. 
1.2. >nat (+, -) / ellipted head noun of genitive Le. an ellipted head noun ofthe genitive is more natura! than a non-ellipted head 
noun ofthe genitive. By the principle ofleast effort; see item (a) in the Introduction. A special case of 1.2: 
1.2.l. >nat ( +/-, -) / ellipted head noun of genitive 
Le. a genitive admitting ellipted and non-ellipted head nouns is more natural than a genitive admitting only non-elliptable head nouns. The scale has the format >nat (A + B, B); see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 

2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1, 1.2.l and 2 it can be deduced: If there is any difference between conversation and the written registers, such 


that one kind of register admits genitives with ellipted and non-ellipted head nouns, and the other kind of register admits only genitives with non-ellipted head nouns, it is conversation that tends to admit genitives with both ellipted and non-ellipted head nouns, and it is the written registers that tend to admit only genitives with non-ellipt­ed head nouns. Q.E.D. 
8. English. Many independent genitives have become conventionalized, so that they need no supporting head noun in the context, e.g. she's going to a friend's. Indepen­dent genitives are found particularlyin conversation (Biber et al. 1999:297, 300). The present deduction continues deduction 7. 
The two syntactic variants: genitive with ellipted head noun, and genitive with non-ellipted head noun. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (+,-)/formula 
Le. formulas are more natural than non-formulas. Formulas support the princi­ple of least effort. See item (a) in the Introduction. 
1.2. >nat ( +, -) / ellipted head noun of genitive 
Le. an ellipted head noun ofthe genitive is more natural than a non-ellipted head noun of the genitive. By the principle ofleast effort; see item (a) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 

2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference between a genitive with ellipted head noun and a geni­


tive with non-ellipted head noun, such that one kind of genitive is used in formulas, and the other not, it is the genitive with ellipted head noun that tends to be used in formulas, and it is the genitive with non-ellipted head noun that tends to be used in non-formulas. Q.E.D. 
9. English. The definite article does not normally combine with the head noun of a double genitive. E.g. a good idea ojJohnny's as against Johnny's good idea (Biber et al. 1999:299). 
The two syntactic variants: the type a good idea ojJohnny's, and the type Johnny's good idea. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (the type Johnny's good idea, the type a good idea ojJohnny's) 
Le. the type Johnny's good idea is more natural than the type a good idea oj Johnny's. By the principle of least effort; see item (a) in the Introduction. 
1.2. >nat (+, -) / definite 
Le. +definite is more natural than -definite. Anything +definite is easier for the speaker to retrieve from memory than anything -definite. By the principle of least effort; see item (a) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference between the type a good idea ojJohnny's and the type 


Johnny's good idea, such that one type is +definite, and the other type is -definite, it is the type Johnny's good idea that tends to be +definite, and it is the type a good idea ojJohnny's that tends to be -definite. Q.E.D. 
10. English. Plural nouns prefer the prepositional genitive to the Saxon genitive. Singular nouns prefer the Saxon genitive to the prepositional one (Biber et al. 1999:303). 
The two syntactic variants: the prepositional and the Saxon genitives. 
l. The assumptions of the Naturalness Theory: 
1.1. >nat (Saxon genitive, prepositional genitive) 
Le. the Saxon genitive is more natural than the prepositional genitive. By the principle of least effort; see item (a) in the Introduction. 
1.2. >nat (singular, plural) 
Le. the singular is more natural than the plural (Mayerthaler 1981:15). By the principle of least effort; see item (a) in the Introduction; while the plural is often encoded by special means, the singular is often left bare (in many languages). 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: If there is any difference between the prepositional and Saxon genitives, such 


that one genitive is used in the singular, and the other genitive is used in the plural, it is the Saxon genitive that tends to occur in the singular, and it is the preposition­al genitive that tends to occur in the plural. Q.E.D. 
11. English. In contradistinction to the prepositional genitive, the Saxon genitive is frequently used in formulaic expressions (collocations) such as life's work,for God's sake, and numerous other combinations with sake (Biber et al. 1999:306). 
The two syntactic variants: the prepositional and the Saxon genitives. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (Saxon genitive, prepositional genitive) 
Le. the Saxon genitive is more natural than the prepositional genitive. By the principle of least effort; see item (a) in the Introduction. 
1.2. >nat (+,-)/formula 
Le. formulas are more natural than non-formulas. Formulas support the princi­ple of least effort. See item (a) in the Introduction. 
A special case of 1.2: 
1.2.l. >nat (+/-,-)/formula 
Le. admitting formulas is more natura! than not admitting them. The scale has the format >nat (A + B, B); see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1, 1.2.1 and 2 it can be deduced: Ifthere is any difference between the prepositional and the Saxon genitives, such 


that one genitive can be used in formulas, and the other genitive cannot, it is the Saxon genitive that tends to admit formulas, and it is the prepositional genitive that tends not to be used in formulas. Q.E.D. 
12. English. The type ajriend ojhers (containing a possessive pronoun, i.e. a pronom­inal noun phrase) is more frequent than the type ajriend ojJohn's (containing a non­pronominal noun phrase) (Biber et al. 1999:308). 
The two syntactic variants: the type ajriend ojJohn's and the type ajriend ojhers. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (the type ajriend ojhers, the type ajriend ojJohn's) 
l.e. the type ajriend ojhers is more natura! than the type ajriend ojJohn's. By the principle of least effort; see item (a) in the Introduction. 
1.2. >nat (more frequent, less frequent) / unit 
Le. a more frequent unit is more natura! than a less frequent unit. See item (e) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: IUhere is any difference between the type ajriend ojJohn's and the type ajriend 


ojhers, such that one type is common, and the other type is less common, it is the type ajriend ojhers that tends to be common, and it is the type ajriend ojJohn's that tends to be less common. Q.E.D. 
(B) Pronouns 
13. English. The personal pronouns of the third person: in the singular, three pro­nouns are used, he, she and it, whereas in the plural only one pronoun is used, name­ly they (Biber et al. 1999:328). 
The two syntactic variants: the personal pronouns of the third person singular, and the personal pronouns of the third person plural. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (three, one)/ number of personal pronouns perone number of the third person 
Le. having three personal pronouns in one number of the third person is more natura! than having only one personal pronoun in one number of the third person. One personal pronoun in one number of the third person is easier for the hearer to process. The situation is the reverse for the speaker; see the Introduction. 
1.2. >nat (singular, plural) 
Le. the singular is more natura! than the plural (Mayerthaler 1981:15). By the principle of least effort; see item (a) in the Introduction; while the plural is often encoded by special means, the singular is often left bare (in many languages). 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: If there is any difference, within the personal pronouns of the third person, 


between the singular and plural, such that one number has three pronouns, and the other number has one pronoun, it is the singular that tends to have three pronouns, and it is the plural that tends to have one pronoun only. Q.E.D. 
14. English. The personal pronouns of the third person singular: he and she have per­sonal reference, and it has mostly non-personal reference (Biber et al. 1999:328). 
The two syntactic variants: the personal pronouns of the third person singular with personal and non-personal reference. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (two, one)/ third-person singular personal pronouns 
Le. having two personal pronouns in the third person singular is more natural than having only one personal pronoun in the third person singular. One personal pronoun in the third person singular is easier for the hearer to process than two per­sonal pronouns in that person. The situation is the reverse for the speaker; see the Introduction. 
1.2. >nat (+,-)/human 
Le. +human is more natura! than -human (Mayerthaler 1981:14). The speaker is more interested in anything human than in anything non-human. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference, within the personal pronouns of the third person sin­


gular, between +human and -human, such that one kind has two pronouns, and the other kind has only one pronoun, it is +human that tends to have two pronouns, and it is -human that tends to have one pronoun only. Q.E.D. 
15. English. In conversation, nominative personal pronouns are often repeated, whereas accusative forms are almost never repeated. E.g. Oh yeah, I. 1 I. 1 really think that you can just go (Biber et al. 1999:334). 
The two syntactic variants: the nominative and the accusative ofthe personal pro­nouns, in conversation. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (nominative, accusative) /in nom.-acc. languages 
Le. the nominative is more natura! than the accusative, in nominative-accusative languages (Mayerthaler 1981:14; 1987:41; Mayerthaler et al. 1998:167). Crosslinguistically, the accusative is sometimes encoded with an adposition, the nominative never. 
1.2. >nat (much, little) / repetition 
Le. much repetition is more natura! than little repetition. Repetition, being imi­tation, is an innate property of human beings (Li 1986:40-1). 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference, in conversation, between the nominative and the accu­


sative of the personal pronouns, such that one kind is often repeated, and the other is hardly repeated, it is the nominative that tends to be repeated, and it is the accu­sative that tends not to be repeated. Q.E.D. 
16. English. In conversation, the expected nominative forms of personal pronouns are replaced by the corresponding accusative forms after BE, as and than, frequently in coordinated noun phrases, and normally in peripheral and non-integrated noun phrases, e.g. it's mg, J'm dozing even better than him, me and this other bloke managed to avoid each other; mg, 1 was early (Biber et al. 1999:335-40). In my opinion, the nominative forms tend to be replaced by the accusative forms when the forms are accented (but not necessarily emphatic). As is well known, the subject pronouns of Germanic languages are as a rule not accented. 
The two syntactic variants: nominative and accusative forms of personal pro­nouns. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (nominative, accusative) /in nom.-acc. languages 
Le. the nominative is more natural than the accusative, in nominative-accusative languages (Mayerthaler 1981:14; 1987:41; Mayerthaler et al. 1998:167). Crosslinguistically, the accusative is sometimes encoded with an adposition, but the nominative never. 
1.2. >nat (-, +) / accented word 
Le. an unaccented word is more natural than an accented word. By the principle of least effort; see item (a) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: If there is any difference between the expected nominative and the accusative 


forms of personal pronouns, such that one kind is used when the pronouns are accented, and the other kind is used when the pronouns are not accented, it is the nominative form that tends to be used when the pronoun is not accented, and it is the accusative form that tends to be used when the pronoun is accented. Q.RD. 
4. Note to item 3. The accusative forms can be replaced by the corresponding reflex­ive pronouns, e.g. Rose and her, also Rose and herself (Biber et al. 1999:343). 
17. 
English. The pronominal possessives have two forms each: a possessive deter­miner, e.g. my, and a possessive pronoun, e.g. mine. An exception is its, which is used as determiner only (Biber et al. 1999:340). This deduction is continued in deduction 

18. 


The two syntactic variants: the determiner and the pronoun use of the pronomi­nal possessives. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (the type my, the type mine) 
Le. the type my is more natural than the type mine. Within pronominal posses­sives, the determiners have less sound body and internal structure than the pro­nouns, therefore they abide by the principle of least effort; see item (a) in the Introduction. 
1.2. >nat (determiner, pronoun) / pronominal possessive 
Le. a pronominal possessive having determiner uses is more natura! than a pronominal possessive having pronoun uses. Determiners are better integrated into the clause (being integrated into a phrase) than pronouns. See item (d) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: If there is any difference, within pronominal possessives, between the type my 


and the type mine, such that one type is used as a pronoun, and the other type is used as a determiner, it is the type my that tends to be used as a determiner, and it is the type mine that tends to be used asa pronoun. Q.E.D. 
18. English. The pronominal possessives have two forms each: a possessive deter­miner, e.g. my, and a possessive pronoun, e.g. mine. An exception is its, which is used as determiner only (Biber et al. 1999:340). This deduction continues deduction 17. 
The two syntactic variants: the acceptable determiner use and unacceptable pro­noun use of the pronominal possessive its. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (determiner, pronoun) / pronominal possessive 
Le. a pronominal possessive having determiner uses is more natural than a pronominal possessive having pronoun uses. Within pronominal possessives, deter­miners have less sound body and interna! structure than pronouns in English, there­fore they abide by the principle of least effort; see item (a) in the Introduction. Moreover, determiners are better integrated into the clause (being integrated into a phrase) than pronouns. See item (d) in the Introduction. 
1.2. >nat (+, -) / acceptable 
Le. being acceptable is more natura! than being unacceptable. See item (i) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference between the determiner use and the pronoun use of the 


pronominal possessive its, such that one use is acceptable, and the other use is not acceptable, it is the determiner use that tends to be acceptable, and it is the pronoun use that tends to be unacceptable. Q.E.D. 
19. English. The emphatic reflexive pronoun mostly stands beside its noun phrase, which is normally the subject, e.g. myself, I don't know. In conversation, the reflexive pronoun stands separated from its noun phrase, usually at the end of the clause, e.g. /'II do it myself (Biber et al. 1999:344). 
The two syntactic variants: the type /'II do it myself and the type myself, I don't know. 
l. The assumptions of Naturalness Theory: 
1.1. >nat (the type /'II do it myself, the type myself, I don't know) 
Le. the type /'II do it myselfis more natural than the type myself, I don't know. The type myself, I don't know is easier for the hearer to process because the reflexive pro­noun and the corresponding noun phrase are contiguous. The situation is the re­verse for the speaker; see the Introduction. Also, myselfis better integrated into its clause in /'II do it myself, and thus the latter is more natural; see item (d) of the Introduction. 
1.2. >nat (conversation, written registers) 
Le. conversation is more natural than the written registers (Dotter 1990, 228). Oral communication is the primary form of communication. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: Ifthere is any difference between the type /'II do it myself and the type myself, I 


don't know, such that one ofthem is used in conversation, and the other in the writ­ten registers, it is the type /'II do it myselfthat tends to occur in conversation, and it is the type myself, I don't know that tends to occur in the written registers. Q.E.D. 
20. English. Demonstrative pronouns normally cannot refer to persons, whereas 
demonstrative determiners, e.g. this man, can (Biber et al. 1999:347). The two syntactic variants: demonstrative pronouns and determiners. 
l. The assumptions of the Naturalness Theory: 
1.1. >nat (determiner, pronoun) / demonstrative 
Le. the determiner use of a demonstrative is more natural than the pronoun use of a demonstrative. Determiners are better integrated into the clause (being inte­grated in to a phrase of the clause) than pronouns. See item ( d) in the Introduction. 
1.2. >nat (+,-)/human referent Le. a human referent is more natural than a non-human referent. This is in line 
with the animacy hierarchy (Croft 1990:112). A special case of 1.2: 
1.2.1. >nat (+/-,-)/human referent 
Le. admitting human referents is more natural than not admitting them. The scale has the format >nat (A + B, B); see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From U, 1.2.1 and 2 it can be deduced: Ifthere i.s any difference between the determiner use of a demonstrative and the 


pronoun use of a demonstrative, such that one use can refer to human referents, and the other use cannot, it is the determiner use that tends to admit human referents, and it is the pronoun use that tends not to admit human referents. Q.E.D. 
21. English. The reciprocal pronoun each other is much more common than the re­ciprocal pronoun one another. Each other is strongly preferred when the reference is to two entities (Biber et al. 1999:346-7). 
The two syntactic variants: the reciprocal pronouns each other and one another. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (each other, one another) / reciprocal pronoun 
Le. the reciprocal pronoun each other is more natural than the reciprocal pro­noun one another. This is based on the relative frequencies of the two pronouns, mentioned above. See item (e) in the Introduction. 
1.2. >nat (two entities, more than two entities) / involved in reciprocity 
Le. only two entities involved in reciprocity is more natural than more than two entities involved in reciprocity. This is based on the real-life circumstance that class­es of two interacting entities are more common than classes of more than two inter­acting entities. See item (e) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1-2 and 2 it can be deduced: If there is any difference between the reciprocal pronouns each other and one 


another, such that one of them is preferred when the reference is to two entities, and the other when the reference is to more than two entities, it is the reciprocal pro­noun each other that tends to be used when the reference is to two entities, and it is the reciprocal pronoun one another that tends to be used when the reference is to more than two entities. Q.E.D. 
22. English. The reference of the demonstrative pronoun to countable entities can be clarified by the addition of one(s), e.g. I like this one. This use is particularly com­mon in conversation (Biber et al. 1999:348). 
The two syntactic variants: the type this/that (referring to countable and uncount­able entities), and the type this/that one (referring to countable entities). 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (the type this/that one, the type this/that + NP) 
Le. the type this/that one is more natural than the type this/that + noun phrase other than one. By the principle of least effort; see item (a) in the Introduction. 
1.2. >nat ( +, -) / countable Le. a countable unit is more natural than an uncountable unit. This is the case 
in languages with i-perspective (Mayerthaler et al. 1998:392). A special case of 1.2: 
1.2.1. >nat (countable, countable & uncountable) 
Le. referring to countable entities only is more natural than referring both to countable and uncoutable entities. The scale has the format >nat (A, A + B); see the Introduction. 
1.3. >nat (conversation, the written registers) Le. conversation is more natural than the written registers (Dotter 1990:228). 
Oral communication is the primary form of communication. A special case of 1.3: 
1.3.1. >nat (conversation, conversation & the written registers) 
Le. occurring in conversation only is more natural than occurring both in con­versation and the written registers. The scale has the format >nat (A, A + B); see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1, 1.2.1, 1.3.1 and 2 it can be deduced: If there is any difference between the type this/that one and the type this/that + 


noun phrase, such that one type refers both to countable and to uncountable enti­ties, and the other type refers only to countable entities, and such that one type occurs in conversation, and the other type occurs both in conversation and the writ­ten registers, it is the type this/that one that tends to refer only to countable entities, and to be used only in conversation, and it is the type this/that + noun phrase that tends to refer both to countable and to uncountable entities, and tends to be used both in conversation and the written registers. Q.E.D. 
23. English. In conversation, the pronoun that is by far the single most common demonstrative pronoun (Biber et al. 1999:349-51). 
The two syntactic variants: the pronoun that in conversation and in the written registers. 
l. The assumptions of Naturalness Theory: 1.1. >nat (vague, precise) / uses of demonstrative pronouns 
Le. vague uses of demonstrative pronouns are more natura! than precise uses of demonstrative pronouns. It is easier for the hearer to process precise uses. The situ­ation is the reverse for the speaker; see the Introduction. 
A special case of 1.1: 
1.1.1. >nat (that, other) / demonstrative pronoun 
Le. the demonstrative pronoun that is more natural than other demonstrative pronouns. In contradistinction to other demonstrative pronouns, the demonstrative pronoun that has vague uses in conversation; it is "usually fairly vague" (Biber et al. 1999:350). 
1.2. >nat (+, -) / frequent 
Le. being frequent is more natural than being less frequent. See item (e) in the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1.1, 1.2 and 2 it can be deduced: If there is any difference, in conversation, between that and other demonstrative 


pronouns, such that one kind of demonstrative pronoun is frequent, and the other kind of demonstrative pronoun less frequent, it is that that tends to be frequent, and it is the other demonstrative pronouns that tend to be less frequent. Q.E.D. 
24. English. There is a special conversational use of the demonstrative determiners this/these, where they introduce new information, especially in telling a story or introducing a new topic, e.g. there was this really good looking bloke (Biber et al. 1999:274). 
The two variants: conversation and the written registers. 
l. The assumptions ofNaturalness Theory: 
1.1. >nat (conversation, the written registers) 
Le. conversation is more natural than the written registers (Dotter 1990:228). Oral communication is the primary form of communication. 
1.2. >nat (old, new) / information 
Le. old information is more natural than new information. Old information is easier for the speaker to handle. By the principle of least effort; see item (a) in the Introduction. 
A special case of 1.2: 
1.2.1. >nat (new, old) / information conveyed by demonstrative determiner THIS. 
Le. new information is more natural than old information if conveyed by the demonstrative determiner THIS. Markedness reversal has been applied to scale 1.2, 
because the environment of the scale was narrowed to the demonstrative determin­er THIS. See the Introduction and item 4. Note below. A special case of 1.2.1: 
1.2.1.1. >nat (new & old, only old) / information conveyed by demonstrative deter­miner THIS. 
Le. new and old information is more natural than only old information if con­veyed by the demonstrative determiner THIS. The scale has the format >nat (A + B, B); see the Introduction. 
2. Markedness agreement (Andersen 2001) applied to naturalness: 
2.1. >nat tends to align with another >nat 
2.2. 
<nat tends to align with another <nat 

3. 
The consequences: From 1.1, 1.2.1.1 and 2 it can be deduced: If there is any difference between conversation and the written registers, such 


that in one kind of register the demonstrative determiner THIS conveys either old or new information, and in the other kind of register the demonstrative determiner THIS conveys only old information, it is in conversation that the demonstrative determin­er THIS tends to convey either old or new information, and it is in' the written regis­ters that the demonstrative determiner THIS tends to convey only old information. 
Q.E.D. 
4. Note. The scale, >nat (THE, THIS) / determiner, shows that THIS is the less natural kind of determiner, by the principle of least effort; see item (a) in the Introduction. 
Conclusion 
In the Consequences of each deduction, a state of affairs is predicted. What is predicted to be such-and-such a state of affairs cannot be otherwise. (In particular, the sta te of affairs is not likely to be the reverse of what it is.) In this sense, each sta te of affairs subsumed in the Consequences is accounted for ("explained" in synchronic terms). 
It can likewise be seen in each deduction which assumptions couched in natu­ralness scales can lead to the corresponding prediction. The creative contribution of the linguist determines which scales are implemented, and in which of the three available scale formats. (In this connection, the essential fact is that the choice ofthe linguist's possibilities is severely limited.) It is conceivable that the same prediction can be deduced from several alternative sets of assumptions. This potential has not been exploited above. 
References 
ANDERSEN, H. 1968: IE. *s after i, u, r, k in Baltic and Slavic. Acta Linguistica Hafniensia 11, 171-190. ANDERSEN, H. 2001: Markedness and the theory oflinguistic change. Andersen, H. (ed.): Actualization: Linguistic change in progress. Amsterdam: Benjamins, 21-57. BIBER, D., S. JoHANSSON, G. LEECH, S. CONRAD & E. FINEGAN 1999: Longman grammar ofspoken and written English. London: Longman. 
CoRBEIT, G. G. 2000: Number. Cambridge: Cambridge University Press. 
CROFT, W. 1990: Typology and universals. Cambridge: Cambridge University Press. DoITER, F. 1990: Nichtarbitraritiit und Ikonizitiit in der Syntax. Hamburg: Buske. DRESSLER, W. U., W. MAYERTHALER, 0. PANAGL & W. U. WURZEL. 1987: Leitmotifs in natura/ morphology. Amsterdam: Benjamins. DRESSLER, W. U. 2000: Naturalness. Booij, G., Chr. Lehmann & J. Mugdan (eds.): Morpho/ogie: ein interna­tionales Handbuch zur Flexion und Wortbildung, Volume I. Berlin: de Gruyter, 288-296. FENK-0CZLON, G. 1991: Frequenz und Kognition -Frequenz und Markiertheit. Folia Linguistica 25, 361-394. HAVERS, W. 1931: Handbuch der erkliirenden Syntax. Heidelberg: Winter. LI, C. N. 1986: Direct and indirect speech: A functional study. Coulmas, F. (ed.): Direct and indirect speech. Berlin: Mouton de Gruyter, 29-45. MAYERTHALER, W. 1981: Morphologische Naturlichkeit. Wiesbaden: Athenaion.-English version: Mayerthaler 1988. MAYERTHALER, W. 1987: System-independent morphological naturalness. Dressler, W. U., W. Mayerthaler, O. Panagl & W. U. Wurzel: Leitmotifs in natura/ morphology. Amsterdam: Benjamins, 25-58. MAYERTHALER, W. 1988: Morphological naturalness. Ann Arbor: Karoma. MAYERTHALER, W. & G. Fliedl. 1993: Natiirlichkeitstheoretische Syntax. Jacobs, J., A. von Stechow, W. Sternefeld & T. Vennemann (eds.): Syntax, Volume I. Berlin: de Gruyter, 610-635. MAYERTHALER, W., G. Fliedl & Chr. Winkler. 1993: Infinitivprominenz in europiiischen Sprachen. Tei/ I: Die Roma­nia (samt Baskisch). Tiibingen: Narr. MAYERTHALER, W., G. Fliedl & Chr. Winkler. 1995: Infinitivprominenz in europiiischen Sprachen. Tei/ II: Der Alpen-Adria-Raum als Schnittstel/e von Germanisch, Romanisch und Slawisch. Tiibingen: Narr. MAYERTHALER, W., G. FLIEDL & Chr. WINKLER. 1998: Lexikon der naturlichkeitstheoretischen Syntax und Morphosyntax. Tiibingen: Stauffenburg. WuRZEL, W. U. 1984: Flexionsmorphologie und Natiirlichkeit. Berlin: Akademie-Verlag. 
Povzetek JEZIKOVNA NARAVNOST V ANGLEŠČINI -(A) RODILNIK, (B) ZAIMKI V Sloveniji smo naravno skladnjo celovške šole razširili na raziskave vedenja sopomenskih in domala sopomenskih (obliko)skladenjskih izrazov, tu imenovanih skladenjske dvojnice. Naše delo je zgoraj ponazorjeno z (obliko)skladnjo angleškega rodilnika in angleških zaimkov. Jezikovno gradivo se obravnava v t.i. izpeljavah. V vsaki izpeljavi je napovedan obstoj nekih (obliko )skladenj­skih razmer, in sicer na podlagi primernih predpostavk in Andersenovih pravil o prirejanju ene vrednosti zaznamovanosti drugi taki vrednosti. Temeljni dosežek teorije: če sta dani dve (obliko)skladenjski dvojnici, od katerih ima ena last­nost A, druga pa lastnost B, zmore teorija odgovoriti na vprašanje, katera izmed obeh dvojnic ima katero izmed lastnosti A in B.