INTO FORMAL SLAVIC LINGUISTICS

PETER KOSTA/JOANNA BlASZCZAK/JENS FRASEKI LJUDMILA GEIST/MARZENA l:YGlS (EDS.) INVESTIGATIONS INTO FORMAL SLAVIC LINGUISTICS CONTRIßUTIONS or THE FOURTH EUROPEAN CONITRENCE ON FORMAL DESCRIPTION or SLAVIC LANGUACES ­ HELD AT POTSDAM UNIVERSITY, NOVEMBER 28­30, 2001 PART 11 ｾ PETER LANG EuropäischerVerlag derWissenschaften FDSL IV Volume 11. V. Syntax 441 KLAUS ABELS: *[P clitic]! ­ Why?................................................................. 443 T ANIA A VGUSTfNOV A: Russian Infinitival Existential Constructions from an HPSG Perspective .. 461 TANIA A VGUSTfNOVA & HANS USZKOREIT: Reconsidering the Relations in Constructions with Non­Verbal Predicates........ 483 LEONARD H. BABBY: Dative Subjects and Nominative Objects: Infinitives in Russian 499 JOANNA BLASZCZAK: Getting Rid of Covert Movement or Getting into Trouble? 517 ZELJKO BOSKOVIC: On Left Branch Extraction 543 STEVEN FRANKS: Case Features, Markedness, and Quantification........................ 579 PETER KOSTA: Adverbs and Negation in Czech 601 SVETLANA KURTES: Genus Verbi in Serbo­Croat AReanalysis of 'se­verbs' 617 VITA G. MARKMAN: On the Syntax and Semantics ofthe Reflexive and Impersonal Passive ­sja ­Verbs in Russian: the Role of Aspect.............. 633 KUNKA MOLLE: The Way ofUsing Articles in Bulgarian Binominative Sentences.......................................................... 651 PETYA ÜSENOVA: On Subject­Verb Agreement in Rulgarian (An HPSG­Based Account) 661 ALLA PASLA WSKA: Negative Polaritätselemente und ihre Lizensierung im Ukrainischen...... 673 HA NA SKRAßALOVA: Comitative Constructions in Czech................................... 685 LUKA SZUCSICH: The Structure of Relative Clauses in Slavic .. 697 YOVKA TISHEVA: Bulgarian yes­no Questions with Partieles nali and nima 715 OLGA MISESKA TOMIC: The Balkan Slavic Future Tenses with Modal Clitics and Tensed Lexical Verbs 731 LUDMILA VESELOVSKA: A Note about Nothing 745 ILSE ZIMMERMANN: The Categorial Dependence of Structural Cases in Russian.... 759 FDSL IV, 601­616 Peter Kosta VI. Semantics 781 TUANA ASIC­KANG'ETHE: The PO, NA & U Opposition in Serbian and Its Equivalents in Some Slavic Languages and Kikuyu 783 LEONlD BIRJULlN: PyccKHe l\HCTjJH6YfHBHble KOHCTPYKl\HH: MexaHH3Mbl pacrrpel\eneHHJI l\ellcTBHJI 797 DAGMAR DIVJAK: An Implementable View on Russian Modificators 815 JENS FRASEK: Polishpewn­, MereoJogy, and Syntax.......................................... 831 EKATERlNA V. RAKHILlNA: Russian Genitive Construction with nomina agentis: Towards a Unified Semantic Description 849 BOZENA ROZWADOWSKA: Initial ßoundary and Telicity in the Semantics ofPerfectivity 859 SERGEI TA TEVOSOV: A Thcory of Slavic Aspect and the Russian Delimitativc 873 DIETER WIRTH: Argumentstrukturelle und sonstige Motive rur den Einschub des kataphorischen Pronomens to (bei bojat 'sja und anderen russischen Verben) 893 Negation and Adverbs in Czech O. Introduction This paper recalls some basic facts on syntax of adverbs and negation in Czech. It has been proposed recently to analyze adverb phrases (AdvP) as the unique specifiers of distinct maximal projections, rather than as adjuncts (ClNQUE 1999). Cinque argues for the existence of a fixed universal hierarchy of clausal functional projections. Furthermore, he rejects the assumption that languages vary in the number and type of functional projections and their distribution. Instead, he tries to construct a plausibility argument against these assumptions, suggesting that no such variation is allowed by UG and that the same number, type and order (hierarchy) of functional projections holds across languages and clause types, despite apparent counterevidence. Specifically, he argues that in addition to the order of free functional morphemes ("particles" and auxiliaries) and of bound functional morphemes (affixes), there is a third type of different classes offunctional projections of AdvP. Similarily to the recent proposal that has been made by Um JUNGHANNS (in press'), I do not adopt the specific approach to the grammar of adverbials that are associated with specific functional projections (cf. also ALEXIADOU 1997 for Greek). Instead, and as opposed to Uwe Junghanns, I make the claim that sentential adverbs are base­generated as adjuncts to an agreement phrase as opposed to VP­Adverbs that adjoin to VP in the base, as shown under (I): (I) Ｏｾ TP T' Subject Ｏｾ T AgrP ｾＯ sentential adverbs (SAs) Agr' ｾ AgrO VP­Adv ｾ VP VP In contrast to my position, Junghanns claims that the three proposed types of sentential adverbs (SA), namely Epistemic SAs (pravdepodobne 'probably', snad 'possibly', 'presumably' ,jiste 'certainly'), Factive SAs (bohuiel 'unfortunately', ovsem 'naturally' , UWE JUNGHANNS (in press) Scope Conllicls Involving Sentential Negation in Czech. In: BROWN, S. & A. PRZEPIÖRKOWSKI (eds.) Negation in Slavic. Bloomington: Indiana: Slavica publishers. 602 FDSLIV poehopitelne 'understandably') and Verificational SAs (opravdu 'in fact, really' samozfejme 'of course', skUlecne 'apparently') "is base generated very low in the structure of the clause, viz. as an adjunct to VP just like any other free (i.e., nonsubcategorized) adverbial". My arguments for at least two different syntactic positions of sentence vs. VP­adverbials is based on the following observation. Whereas SAs usually and mostly do not affect the reading of the negated proposition despite their apparent scope conflicts sharply described and explained in the mentioned article by UWE JUNGHANNS (in press), manner adverbs such as pekne 'nicely', vUdne 'moderately', nahlas 'loudly', dohre 'good', etc. strictly combine with partial negation with different scope properties than the sentence negation has. The following paper is organized as folIows: in the reminder of part one I shall present some evidence for the position of the sentential negation in Czech. In the second part I shall show how sentential negation and sentential adverbs interact and how scope conflicts can be resolved under an analysis of covert movement. In the third part of my paper, some evidence of the distribution of manner adverbs and negation will be presented. The conclusion of this paper will be an analysis of covert movement of the finite verb out of the scope domain of the constituent negation (focus negation) leaving the manner adverb in situ under c­command. 1. Sentential Negation, eonstituent negation and adverbs in ezeeh It has been claimed in many recent publications that the freedom of word order of languages like Czech is not as fTee as it seems. There are restrictions on word order that result either from requirements of Syntax, Phonology or Semantics. My analysis is based on the assumption that the syntactic position of the surface word order of arguments and adjuncts reflects (also in ist relation to sentence prosody) the categorial representation of the cognitive meaning of the sentence. Following I lAJICOV A (I 995a,b), it seems necessary to distinguish between the focus of a rhematizcr, the focus position of the adjuncts (adverbials) and the focus of the arguments. In addition to it, it seems that focus position and scope properties of sentential negation vs. constituent negation are closely connected with the focus position and scope properties of scntential vs. VPAdverbs as scope taking items. Consider the following example (2): (2) [AGRSpUreite [AGRS!' AgrSO jsem [11' [NEGP <nei'ekla;> [AGROr nie [FOC VP tj [VI' aUXlsg Neg told nothingAKK Surely NAhlas])])]ll. loudly "It is sure that I haven't said anything loudly." The first adverbial on the leftmost edge of the sentence is a SA of the verificational or epistemic class, thc last adverbial on the right periphery of the sentenee is a manner adverb in a lower adverb position. 2 Now, consider the differences in the foeus and seope Benjamin Shaer (2000) shows that some manner adverbs such as rudely exhibit different readings depending on their syntaetie position. In a sentenee (2a) Louisa rudely answered Patricia the reading ean be construed as saying that Louisa's rudeness consisted in her having answered Patricia whereas in (2b) Louisa answered Patricia rudely loeates the flaw in the manner of P. KOSTA, Negation and Adverbs in Czech 603 domains of the two classes of adverbs. The SA c­commands the negation, the NegP and the NegV complex c­command the rest of the sentence (Le. the direct object argument nie 'nothing' and the manner adverb nahlas 'Ioudly'), the resulting reading is ambiguous (2a) vs. (2b): (2a) "it is sure that anything that has been said did not happen in a loudly manner" (2b) "it is sure that nothing has been said loudly" The first interpretation a1lows apresupposition on the existenee of the asserti ve meaning of the proposition (finite verb) of "saying, but not anything loudly" whereas the interpretation (2b) excludes the speech aet of saying anything at all, Le. the verb and all its subcategorized elements (direet object argument, including the manner adverb) are negated. This ambiguity does not emerge with the SA due to its higher position above the sentential negation. LANG (1979, 1983) makes the point that SAs semantieally outscope sentential negation, i.e. they are not interpreted within the seope of another funetor. This property of sentential adverbs, however, does not hold for manner adverbs of the lower class (presumably complements in a VP shell in the sense of LARSON (1988)) as seen in (2) and (3a­d): (3) a. ze jsem pn tom <nevypadala> zdaleka tak pekne jako Rhonda far as nieely as Rhonda that aUXlsg thereby Neglooked "that I haven't by far looked so nieely Iike Rhonda" b. Tyhle konciny <nepusobily> zrovna v/(dne. these sOIToundings Negimpress speeifieallyeomfortably/moderately "These sOIToundings did not impress by being very eomfortable" c. <Nereagovala> pfflis dohre. Neg reaet tao good "She did not reaet very good" d. Pfflezitostnf kufaei totiz <neinhaluji> tak laene Occasional smokers namely Neg inhale that avidly "Oeeasional smokers do not inhale that mueh" The meaning of (3a) includs the presupposition of "looking nieely but by far not that nice as R." where Neg takes a contrast focus reading over the manner adverb and does not attract the negation of the verb (Le., it outscopes the finite verb), resulting in the interpretation of constituent negation of the manner adverb. The meaning of the manner adverb in Ob) is also negative bearing contrast focus as opposed to some more nice parts of the country that are not mentioned in the text. The interpretation of (3c) is that she reacted but not in a very good manner. The interpretation of (3d) is of course not that occasional smokers do not inhale at a1l but that they da not do it that avidly. The above mentioned examples with manner adverbs seem to allow a first descriptive answering. Manner adverbs ofthe lower dass are usually subcategorized for verbs that take adverb complements excluding the higher (=2a) position, cf. Joan behaved rudely ff • Joan rudely behaved. In Czech, however, the contrast of the two positions of manner adverbs results in the contrast between focussed reading vs. unrnarked (unfocussed) reading; cf. Petr se choval neukO.zneni!/sproste ff Petr se neukO.zneneJsproste CHOVAL. We will keep these contrasts in mind for the sake ofthe main claim of our study (c( Section 3). FDSLIV 604 generalization, namely (4): (4) First decriptive generalization on manner adverbs and negation: Manner adverbs take scope of negation as if it were the constituent negation Let us reconsider some examples with SAs confirming that the descriptive generalization (4) does not include anything else than this c1ass of adverbs: (5) [Tenkrat se [clovek [vlastm?" [nikdejinde] actually V­SA nowhere else LOCALADVERD At that timeTIMEADVERB REFL manNOM ani [usadit [<nemohl>]]]]] . evenNEGPART setde down NEG could "At that time none actually could have settled down at any other placc" The example (5) shows the designated syntactic positions of different c1asses of adverbs having the preference order time adverb (tenkrat) ­ SA (V­SA) ­ location adverb outscoping the time adverb and the Verificational SA out of the sentential negation but including the negation of the local adverb in front of the modal (matrix) verb. The unmarked word order of sentential and manner adverbs 3 in Czech sentence can be confirmed by some examples with negated and affirmative Wh­phrases; cf. (6): (6) a jak dlouho. promlouval, ale nikdo nikdy <nevldel> kdy. kde (He) spoke but noone never Negknew when where and how long "He was speaking but none ever knew when where and how long it would be" Subject> Time> Neg > V Time> Location> Manncr Notice that the scope contrasts between negation, manner adverbs and SA also account for the difference between the semantic c1ass of subject­oriented adverbs which behave syntactically like real SA on one hand and manner adverbs on the other; cf.: (7) Elenino tela <nevystavili> zamerne. bodYAKK Neg exhibited on purpose E.s "They did it on purpose (it was their purpose) NOT to give the body of E. to thc exhibition" Even though the subject­oriented adverb zamerne "on purpose" lands in a surface position that corresponds to the designated position of manner adverbs (cf. 2, 3a­d), it shows different syntactic behavior as to the scope properties of the negation, excluding the reading of constituent negation of manner adverbs. Now, reconsider some problems c10se connected with the observation made by Uwe Junghanns that even SAs in Czech may enter a syntactic domain where negation precedes a sentential adverb so that, on the surface, the relative scope of the two items is as shown in (8)­(10) [Junghanns in press:2, examples (I )­(2)]: is obvious that the adverbs of both groups divide into different semantic subclasses. THOMAS ERNST (2000:80) proposes two groups of adverbs, namely functional adverbs and predicational adverbs, first being a dose dass, latter being an open dass. For the purpose of our study, however, the distinction between SA and manner adverbs suffices because we do not concentrate on other types of adverbials. It P. KOSTA, Negation and Adverbs in Czech 605 (8) Ja [ne]pjsem [samoztejme]av situaci, ze bych si mohl INoM [Neg]p am'P/SG [of course]ain situation that auxlP/SO refl DAT could vybirat [...] chOOSelNF "Of course, I don 't have a choice" (9) Tento ideologicky pseudoproblem nebyl bohuiel This ideological pseudo­problemNoM [Neg] ll...wasJP/so [unfortunately] a jedinYm nasledkem povedene oslavy skolnikovych padesatin. only consequencelNSTR held ceiebratioo"EN of the pupils' fiftieth­anniversary (JUNGHANNS, in press:2) (10) [Neg [SA [... This is an apparent counterevidence to the semantic intuition according to which SAs should take scope over Neg as illustrated in (11) and as demonstrated under (2,5, 7): (11) [SA [Neg [.... ]]] Junghanns correctly points at the fact that the surface configuration does not reflect the correct scope relation, thus, proposing non­overt­movement that creates a configuration in which the item that semantically outscopes the other one c­commands it. That means, that semantic scope is determined by syntactic hierarchy, but the surface does not necessarily reflect it. Scope conflicts are overcome after Speil­Out, at an abstract, postsurface, level of representation (LF). The offending item moves to a position outside the scope of sentential negation. This approach is reminiscent of May's (1985) Quantifier Raising (QR) at LF. Dut whereas MA Y (1985) introduces QR to account for scope ambiguities, Junghanns deals with cases in Czech where there is no ambiguity at all. For the Czech examples (8)­(9) Junghanns proposes a non­overt movement analysis out of the scope of the sentential negation. 2. Sentential Negation and sentential adverbs In the following section I try to reject the proposal by UWE JUNGHANNS (in press) stating that sentential adverbs are base­generated very low in the structure of the clause, viz. as an adjunct to VP just like any other free (i.e., non­subcategorized) adverbial. Instead, I make the claim that at least some aspects of adverb distribution can be predicted from their semantics and that the mapping between syntax and semantics ought to be as direct as possible (cf. ERNST 2000:81). Junghanns gives three arguments in favor of a base­generated position adjoined to VP. The first piece of evidence comes from the periphrastic future. Under the assumption that the future auxiliary is generated in the functional head T(ense), in example (12) the SA follows aux. The conclusion that SA must be lower in the structure than T is not ｾ straightforward as one might suggest. Let us first reconsider the structure proposed by Junghanns: 606 FDSLIV P. KOSTA, Negation and Adverbs in Czech 607 (12) (op Takorymi a podobnymi nametl'] se [TP [T budou] (AGRap tREfL [vp theseINsTR and similarlNSTR tOpiCSINSTR refl aux3PlPL (nepochybne]a [vp (vp tNOM zabYvat tINSTR] [OP navrhovatele nasich [no­doubt]a deal'Nf person­who­submitNoMIPL our novych ustav JJJ]] new constitutions "No doubt, those who will submit the drafts of our new constitutions will be concemed with these and similar topics" (JUNGHANNS, in press: 11, ｾ ex. 17ab) (here to support the c1itic auxiliary jsem), aux moves to the head of CO and the SA remains in its base­generated position, namely left adjoined to AGRSP; cf.: Under the assurnption that the future aux has not moved out of its base­generated position in TO Junghanns is correct. On Cinque's approach, however, each adverb is licensed in a one­to­one relation with a specific functional head having a relatcd meaning, and these heads are ordered by UG in a rigid clausal hierarchy. Thus for (12) the relevant portion ofthe clause would look something like (13) (the actual node labels are not important at this point): The designated position for SA of the epistemic class in Czech is confinned by data from Russian with pied piping of the adverbial adjunct together with the participal adjunct into a position either leftadjoined to or above AGRSP, presurnably into SpecCP as demonstrated under (14) (cf. KOSTA 1998: 148, ex. 7): (13) TenseP ｾ Tense bulau Epist' ＧＯｾｍ nepoehybne Eplst I odP m]J. (15) ... se ... ozvalld'ik decka, [ep ktere [xp patme [xp upadlo]]J (KOSTA 1998:145) was to be heard the voice of a child that presumably fallen has ｍｏ､ｾａ｢ｩｉｐ ａ｢ｩｬＯｾ VP VP ｾＯ zabtat vP (Marmer Adverbs] dükLdne In the Cinque­style adverbial syntax (13) the epistemic adverb nepoehybne stands above a modal node but undemeath a tense node under which the future aux is base generated. The structure (12) can be derived. But there is a clear counterevidence in Czech that SAs of the epistemic class must remain in situ in SpecEpistP (of the base­generated EpistP). Recall the facts in (2), where an epistemic SAs can be adjoined to any maximal projection in Czech, including AGRSP in (2), here repeated as (2'): (2') (14) [cP [cP [AP verojatno upavsi]. [vp tj [vp zakrical [OPI REBENOKi t. Presumably having fallenPart started to cry a child/boy "A boy presurnably having fallen started to cry" In the Czech translation of this sentence from Anna Karenina the SA verojatno remains in situ in a position before the main verb in a restrictive relative clause, by which the proposition by the speaker is judged as true based on how the real world situation, or source of knowledge, affects the likelihood that the corresponding event occurred. It is not clear, however, which functional projection the SA adjoines to: 'P EPlst ａ､ｶｾ (2") [CP Ja CO jsem; [AGRSP urCite [AGRSP AgrSO t; [TP [NEGP <nefekla> [AGRaP nie [VP t aUXlsg Neg told nothingAKK Surely [VP NAh/as ]mJJ]. loudly "It is sure that it was ME who hasn't said anything loudly" (AGRSP UrCite [AGRSP PRO AgrSO jsem [TP [NEGP <nefekla> [AGRaP nie [YP t [VP aUXlsg Neg told nothingAJ(K Surely nah/as lllJlll loudly "It is sure that I haven't said anything loudly" Exarnple (2) and the structure (2') demonstrate the Null­Subject version of this sentence, having a topicalized version with lexical subject (2"). My proposal for the latter structure is that the lexical subject has moved to SpecCP for independent reasons As already stated in the beginning of my paper, I do not share the oppinion that specific functional projections are designated for specific SAs, but I do recall my assumption that in structures like (12) the position of the future aux is not clear. In other wards, it is not clear whether the future aux has moved or not. Under the assumption that the reflexive particle se must have moved to a position designated for clitics in Czech (the position must be the Wackemagel, i.e 2nd position) and assurning that the DP [op Takovymi a podobnymi namety] has been topicalized in (12), we can still claim that the structure (12) is something like (12'), where the epistemic SA is right adjoined to AGRSP, cf. (12'): O (12') [CP [CP [op Takovymi a podobnYmi nametl'] [CO sei] [AGRSP AGR budouj [AGRSP nepochybne] [TP tj [ [AGROP tj [VP zab)'vat [VP [DP tNDM t'NSTR] [op navrhovatele naSich novych ustav]]JJJ My proposal for the sentence structure including SA of the epistemic c1ass is also confinned by the following empirical facts: • The Uflffiarked word order shows up with SAs preceding the NegP including second position pronominal c1itics (here the pronoun ho "hirn" CL AKK ) that follow the epistemic adverb. The marmer adverb (here the negated marmer adverb nijak 'in no way') is included into the scope of negation. The scope of Negation in such sentences is narrow, i.e. the negation overtly outscopes the SAs, cf. (16). 1608 FDSLIV P. KOSTA, Negation and Adverbs in Czech Preposed (scrambled) or topicalized argument DPs usually precede SAs and still are in the scope of sentential negation, whereas the SA is outscoped. The negated finite verb is mostly adjacent to the SA and to the direct object DP(cf. 17­19). In sentences with double objects the prepositional phrase precedes and the direct object DP follows the verb (cf. 18). In structures with negated existential SC the SA has scope over an event of the matrix clause (here marked with the demonstrativ tohle "this") but is outscoped by the Negation of the SC (cf. 20): If we replace the epistemic SA urcite and put it to the right periphery position for which the manner adverbs are subcategorized we get a quirky interpretation that the way of saying something happened NOT in a "determined, resolute" tone; cf.: (2" ') [Nahlas [AGRSP AgrSO jsem [TP [NEGP <nefekla­> [AGROP nie [VP t [rocVP URCite ]]JJ)]] Loudly aUXlsg Neg told nothingAKK "Loudly I haven't said anything in a detennined way/resolutely" (without comma intonation) This is to say, that the syntactic reordering of the different classes of adverbs leads to a semantic effect which seems to support our view that the mapping between syntax and semantics ought to be as direct as possible. Additionally, the changed status of the adverbs is confirmed by the fact that the negation does not outscope the manner adverb urCite "in a determined way, resolutely". I should stress that prosody is a very important point of further research, because the same structure changes the meaning with respect to different prosodical status of the structure. With comma intonation or after a break the adverb urCite obtains the semantics of an epistemic SA again: O (2'''') [Nahlas [AG'SP AgrS jsem [TP [NEGP <nefekla> [AüROpnic [VP t # [VP URCite J)]lJl) Loudly aUXlsg Neg told nothingAKK "Loudly I haven't said anything, that's for sure" (# is a prosodical break after a comma intonation) nijak <nelakala> pi'edstava, ze by se musei, (16) Rozhodne ho absolutely himAKKCL noway Neg attracted imagination, that he had to, vydechla Joanna sihted Joanna "Absolutely, the vision did not attract hirn, that he had to, said (sighed) Joanna" (17) iddnym toxickym lätkäm. ureite <neholdovala> nODATIPL toxiCDATIPL substanceSDATIPL surcly (she) Ncg prefered "Surely, she did not prefer any toxic substances." (18) Na Joannu ocividne zvlastni dojem <neudelala> 1'0 JoannaAKK obviously a special impressionAKK Neg made "Obviously, Oll Joanna she did not make any special impression" (19) Ve sklepe urCite <nehoi'elo> In the cellarLoc/sG surely Neg bumcd "Surely, it did not bum in the cellar" (20) Tohle urcite <neni> doktorka Blalockova This surely Neg iS3P/sG doctor BlalockovaNOM/SG "Surely, this is not doctor Blalockova" All above mentioned sentences can be derived by the same mechanism. The purpose of my proposal was to show that there is a fixed base­generated position of epistemic SAs in Czech (presumably left­ or right­adjoined to AGRSP) and that the word order variation can be derived either by scrambling of argument phrases or by topicalization both being unbound processes that feed the information structure of the sentence. It seems to me that a derivational theory on movement has to respect the fact that rightward movement is banned (cf. LCA in KA YNE 1995) and the overt movement of the fmite verb has to follow principals of economy. Our proposal on sentence structure ofSAs vs. manner adverbs and the generalization (4) should predict that (a) SAs should be banned from a right periphery (let us call it the focus, = FOC) position, and (b) if they appear there they should persist scope conflict effects by covert movement on LF. Let us first reconsider some facts on the distribution of SAs of the epistemic, factive and verificational c1asscs on onc hand and the manner adverbs on the other hand: recall the following examples: (2) [Ureite [AGRSP AgrSO jsem [rp [NEGP <nei'ekla> [AGROpnie [YP t [FOCVP NAhlas]]})]JJ Surely aUXI,g Neg told nothingAKK loudly "It is sure that I haven't said anything loudly" 609 The possibility of reconstruction manner adverbs into SA is confirmed by the fact that the SA persists scope conmcl effects with the sentential negation by covert movement on LF to a higher c­commanding position above the NegP. Needless to say, that the manner adverb reading of ureite is only obtainable with certain verbs for which this adverb is subcategorized. In our texlcorpus (taken from the Czech National Corpus UCNK) most verbs exclude the manner reading of the SAs, disal10wing to place them in the right­adjoined position as complement of a VP. We can demonstrate this on the following examplcs: (5') ) ) [Tenkrat se [clovek [t [nikdejinde) ani nowhere else LocAL ADVERB evenNEGPART At that time­nME ADVERB REFL manNOM [vlastne)o.k. [usadit [vlastne]o.k. [<nemohl>] *[vlaslneJJ]]]. settle down NEG could actually V­SA "At that time none could have settled down at any other place *actually" In (5') the lowest right adjoined position is excluded whereas the position adjacent to the verb complex or intervening between the negated modal and the main verb is possible (wilh SA­ reading). (7') Elenino telo <nevystavili> zamerne. E.s bodYAKK Neg exhibited on purpose "They did it on purpose (it was their purpose) NOT to give the body of E. to the exhibition" *''The way or manner of exhibiting the body was not on purpose" FDSLIV 610 In (7') the maner reading is excluded because zamerne is a subject­oriented SA and the verb "vystavitJvystavovat" does not subcategorize for such an adverb. (17') +iddnym toxick)rm latkfun (18') (19') (20') (21) (22) 3. <neholdovala> urCite nODATIPL toxiCDATIPL substanceSDATIPL (she) Neg preferred surely "+She did not prefer any toxic substances in a detennined way." +Na Joannu zvlastni dojem <neudelala:> oCividne obviously To JoannaAKK a special impressionAKK Neg made "+On Joanna she did not make any special impression in an obvious way" *Ve sk1epe <nehorelo> urcite In the cellarLoc/SG Neg bumed surely "*it did not bUffi in the cellar in a detennined way" *Tohle <neni> doktorka Blalockova urCite This Neg be doctor Blalockova surely "This is not doctor Blalockova in a detennined way" bez proste <neexistovala> II *proste. neho by parta II *simply Without rum aUXCOND companYNoM simply Neg existed "Surely, without hirn the company would not exist"I??... would not exist in a simple way" ｭ･ｾＮｶｯ takhle fict 11* ovsem, ponevadz.. .. <Nemohl> to Neg could this however that way say II *however, because... "He could not, however, say ist that way" Constituent Negation or sentential negation and VP­Adverbs (manner adverbs) The analysis suggested for the syntax of adverbs is based on the descriptive generalization under (4), here repeated as (4'): (4') Second descriptive generalization on manner adverbs and negation: First descriptive generalization + manner adverbs are subcategorized by the verb The contrast of manner adverbs with respect to structural position and the SA is striking both with respect to lexicon (semantics) and to scope properties of negation. Recall the following examp1es with manner adverbs: (3') a.... ze jsem pi'i tom <nevypadala:> zdaleka takpekne jako Rhonda far as nicely as Rhonda that aUXls g thereby Neglooked "that I haven't by far looked so nicely like Rhonda" b. TyWe konciny <nepusobily> zrovna vlEdne these sorroundings Negimpress specifically comfortably/moderately "These sorroundings did not impress by being very comfortab1e" c. <Nereagovala> prilis dobfe Neg react too good "She did not react very good" d. Pfilezitostni kuraci totiz <neinhaluji> tak laene Occasional smokers namely Neg inhale that avidly "Occasional smokers do not inhale that avidly" P. KOSTA, Negation and Adverbs in Czech 611 1n all these sentences the negation reaches only the manner adverb and outscopes the verb in contrast to sentences with SA where the negation reaches only the verb and outscopes the SA. Bearing these facts in mind, there is a close connection with the analysis proposed by Uwe Junghanns and in the remainder of my paper I shall try to explain the different behavior of SA and manner adverbs and negation. 4. On structure 0/ negation, VP­adverbs and covert V­movement I now turn to thc question raised in the beginning of my paper, namely, why do manner adverbs stay in the scope of Neg and why are SAs outscoped by sentential negation? My proposal will be the following: The fact that in most cases SAs do not meet scope conflicts (for the exceptions see JUNGHANNS, in press) can be explained by the fact that sentential negation does not build a scope domain of SA, the latter being situated higher in the sentence structure. Whereas SAs usually and mostly do not affect the reading of the negated proposition ­ despite their apparent scope conflicts that are then resolved by covered movement after Speil­out out of the scope of sentential negation (cf. JUNGHANNS in press) ­ manner adverbs such as pekne 'nicely', vlEdne 'moderately', nah/as 'loudly', dobfe 'good', etc. strictly combine with constituent negation interpretations with different scope properties than the sentence negation has. As I already stated my proposal will be that the negation ­ besides its main property as operator responsible for binding the trace or variable of negated sentences ­ constitutes the focus domain of the lowest phrase, viz. VP­shell. The starting point of my analysis is then the assumption that the negation ­like other scope taking items (e.g., the particle only, etc.) ­ can be a candidate for focalizers or focus sensitive particles in the sense of HAJlcovA (l995a,b). I assume that the focus feature [FOCJ can be assigned to syntactic constituents that include SAs and manner adverbs. Because SAs must take scope over the whole proposition (they take the proposition as their argument in complement position) they have to stand higher than the NegP. Scope conflicts between a sentential negation and SAs as discussed in Junghanns are resolved or overcome in the non­overt part of syntax (traditionally called LF). After Speil­out, the SA, the offending item, leaves its base position to adjoin to CP (or in my version to AGRSP) from where it takes scope over the whole clause. In the abstract representation the correct relative scope is detennined as folIows: (23) Scope resolution after Speil­Out: The SA moves to adjoin to CP [cp [SA] [CP... [NEGP [Neg] .... [vp tSA [VP... ]]]]] (Junghanns, in press) The question that arises ｡ｾ far as to manner adverbs and negation is now: Why do not manner adverbs do the same as SAs and leave their base position by overt or covert movement? We have a solution consisting oftwo proposals: First proposal: the negation in sentences with manner adverbs behaves Iike a focus particle marking contrast focus as constituent negation. But it is areal sentence negation. This assumption forces us to make a new proposal on the structure of the VP containing manner adverbs. FDSLIV 612 • Second proposal: adjuncts are islands for movement, altematively inner negation islands are sensitive for manner adverbs as real adjuncts only. Let us first start with the second proposal: this solution is conceptually not very attractive because of some contradictions conceming the theoretical status of tests for long Wh­movement out ofadjunct clauses and inner islands anyway. Let us repeat some of these problems. As a typical case of violation of the ECP the Wh­extraction out of negated adjunct islands has to be mentioned: (24) a. Komu jsi nefekJ t pravdu? WhomDAT aux Negtold truth AKK "To whom didn't you tell the truth?" b. *Jak jsi se nechoval ti? *How didn't you behave ti? It seems to be more a question of semantics or pragmatics whether we can ask such questions. This can be confirmed by the fact that we can answer questions such as (3") but in English and Czech the question would be more adequate as positive question and the answer would be consistent if we negated the manner adverb and not the verb: (3") a. Jak jsem pi'i tom <vypadala>? "How did I thereby look?" b. Jak tyhle konciny <püsobily>? "How did these surroundings appear?" c. Jak <reagovala>? "How did she react?" d. Jak pfileiitostnf kui'aci totiz <inhalujf>? "How do occasional smokers inhale?" ­ Zdaleka ne tak pekne jako Rhonda. "By far not as nice as Rhonda" ­ Ne zrovna vlidne. "Not quite comfortable" ­ Ne pi'ilis dohre. "Not very good" ­ Ne tak lacne. "Not that avidly" This kind of evidence supports the first proposal, namely that negation in sentences with manner adverbs is a focus particle marking contrast focus as constituent negation. The homonymy of sentential negation and constituen negation in Czech is resolved after Speil­out as covert movement of the finite verb out of the scope of the focus particle because the constituent negation can have scope just over one constituent in the phrase. This assumption forces us to make a new proposal on the structure ofthe VP containing manner adverbs. My proposal on the structure of manner adjuncts is based on the work by PROGOVAC (1998) who has shown that certain adverbs can be analyzed as predicates of events, as illustrated in (26) for (25). In English, an overt conjunction can introduce a manner adverb, as in (27). In Serbocroatian, both an overt conjunction and what looks like an event pronoun/demonstrative can surface, as illustrated in (28): (25) (26) (27) (28) John read the book avidly. 3e (RG,b,e) & (A,e)) John read the book and avidly. Jovan je proCitao knjigu, and is read book John to iedno. that thirstily (PROGOVAC 1998:258) P. KüSTA, Negation and Adverbs in Czech 613 In SC the overt conjunction with a manner adverb is necessarily accompanied by what seems to be an overt counterpart of the event variable e, which is exactIy in the right position to serve as an argument of the predicate avidly. In the same fashion we can argue that manner adverbs are predicates of a complex VP coordinated with the main verb (and its arguments). Because of the ban on overt conjunctions (formulated for appeal to Economy as a principle of grammar, see Chomsky, as 'Avoid Conjunction' principle) before Spell­out the conjunction has to be deleted (possibly at PF). If a sentential negation c­commands the coordinated complex VP, it must overtly merge via operation ATTRACT with the next available predicate which is the verb of the higher VP. Since the manner predicate inside of the light vP is the right adjoined sister to VP, it stands in a lower (complement) position, and thus cannot be attracted to the O head of the Neg first. This is the overt part of the syntax. But semantically, the both predicates are coordinated with the event variable e, so that the scope of negation reaches both parts of the complex predicate. For independet reasons the upper predicate has to raise covertly out of the focus domain of its base position leaving a trace that is still headed by the complex Neg­V. The reasons for this cover! movement at LF might be the same as proposed by UWE JUNGHANNS (in press), namely becoming a part of background or topic of the sentence and outscoping out of a focus domain in the sense of HAllcovA (1999a,b). The remaining part of the complex VP gets the focussed reading, optionaly also a contrast focus reading. (29) [UrCite [AGRSP AgrSOjsem[ TP [NEGP <nei'ekla;> [AGROP nie [FOeVP t; [&P to [vP Surely aUxlsg Neg told nothingAKK NAhlas]]J]]]]. loudly "It is sure that I haven't said anything (and) loudly" NegP (30) Ｏｾ NegP' SpecNeg mc Ｏｾ O &P [=FOC] Neg nei'ekla ｾＯ vP [FOC] ｾ VP e (to) ｾ nahlas FDSLIV 614 P. KOSTA, Negation and Adverbs in Czech BROWN, S. & S. FRANKS (1995) Asymmetries in the Scope of Russian Negation. In: Journal AGROP (31) nlC Ｏｾｎ 615 01 Slavic Linguistics 3(2):239­287. '1 ' eg P2 = sententla negatIOn Ｏｾ VP nefekla ｾＯ fekla NegP I = constituent negation ne Ｏｾ ｾ vP nahlas Another alternative would have to inc1ude the cyclical operation phase by phase. With the new version ofthe Minimalist Program ­ the Minimalist Inquiries (CHOMSKY 1998) _ the distinction between overt and covert movement collapse. Instead, all operations must apply cyc1ically. In the case of manner adverb this means that it must have its negative feature checked by the constituent negation marker against the little v within vP since vP, being a phase, is subject to the following cyc1icity condition: "The head of aphase it 'inert' after the phase is completed, triggering no further operations" (p. 20). This in turn predicts that it should be impossible for the verb of the VP to have its negativy feature checked later in the derivation than at the vP level. This new mechanism would have the advantage that we can dispense with a covert movement of the verb up to NegP because we get the interpretation of constituent negation of the manner adverb for free without necessarily ascribing the verb nefekla negative mcaning, cf. (3 I). This account also explains why constituent negation merges with the verb overtly assigning the manner adverb negative meaning and exc1uding the negation of thc clause. Bibliography ADGER, D. & G. TSOULAS (2000) Aspect and Lower VP Adverbials. In: ALEXIAOOU, A. & P. SVENONnJS (eds.) Adverbs and Adjunction. Ｈｾ Linguistics in Potsdam, 6: 1­18). Potsdam. ALEXlAOOU, A. & P. SVENONIUS (eds.) (2000) Adverbs and Adjunction. Ｈｾ Linguistics in Potsdam, 6). ｭ｡､ｾｴｯｐ ALEXlADOU, A. (1997) Adverb Placement: A Case Study in Antisymmetric Syntax. Ｈｾ Linguistics today, 18). Amsterdam: lohn Benjamins. BABBY, L. H. (1980) Existential Sentences and Negation in Russian Ann Arbor. BARTSCH, R. (1972) Adverbialsemantik. Die Konstitution logisch­semantischer Repräsentationen von Forschungen, 6). Frankfurt a.M.: Athenäum. Adverbialkonstruktionen Ｈｾｌｩｮｧｵ ｳｴｩ ｣ｨ･ BELLERT, I. (1977) On Semantic and Distributional Properties of Sentential Adverbs. In: Linguistic Inquiry 8(2):337­350. BLASZCZAK, 1. (2000) Irrvestigation into the Interaction between Indefinites and Negation in Polish. Doctoral dissertation, HUB, Berlin. BROWN, S. (1996) The Syntax 01 Negation in Russian Docloral dissertation, Indiana UniversilY· BROWN, S. & S. FRANKS (1997) The Syntax ofPleonastic Negation in Russian. In: Formal Approaches to Slavic Linguistics 4. The Cornel/ Meeting 1995, 135­164. Ann Arbor: Michigan Slavic Publications. CHOMSKY, N. (1995) The Minimalist Program. Cambridge, MA: MIT Press. CHOMSKY, NOAM (2000) Minimalist Inquiries: The Framework. In: MARTIN, ROGER, DAVID MICHAELS & lUAN URIAGEREKA (eds.) Step by Step. Essays on Minimalism in Honor 01 Ho ward Lasnik, 89­155. Cambridge, MA: MIT Press. CINQUE. G. (1999) Adverbs and Functional Heaels.· A Crosslinguistic Perspective. New York: Oxford Univ. Press. DoWTY, D. (2000) The Dual Analysis of Adjuncts/Complements in Categorial Grammar. In: FADRICIUSHANSEN, c., LANG, E. & C. MAIENBORN (eds.) Approaching the Grammar 01 Adjuncts. Proceedings olthe Oslo Conlerence. September 22­25. 1999. ｓａｚｾＨ Papers in Linguistics, 17:53­78). ENDRISS. C. (200 I) The Double Scope olQuantijier Phrases. Potsdam Ｈｾｌｩｮｧｵｩｳｴ ｣ｳ in Potsdam, 17). ERNST, TH. (2000a) Semantic Features and the Distribution of Adverbs. In: FABRICIUS­HANSEN, c.. E. LANG & C. MAIENDORN (eds.) Approaching the Grammar 01 Adjuncts. Proceedings 01 the Oslo Papers in Linguistics, 17:79­98). Conlerence. September 22­25. 1999. ｓａｚｾＨ ERNST, TH. (2000b) On the Order of Event­Intemal Adjuncts. In: ALEXIADOU, A. & P. SVENONIUS (eds.) Adverbs and Adjunction. Ｈｾ Linguistics in Potsdam, 6:33­49). Potsdam. FREY, W. (2000) Syntactic Requirements on Adjuncts. In: FADRICIUS­HANSEN, c., E. LANG & C. MAIENBORN (eds.) Approaching the Grammar 01 Adjuncts. Proceedings of the Oslo Conference, September 22­25, 1999. ZAS Papers in Linguistics 17: 107­134. HAEGEMAN, L. (1995): The Syntax 01 Negation. Cambridge: Cambridge Uni. Press. HAIDER, H. (2000) Adverb Placement ­ Convergence of Strueture and Licensing. In: ALEXlADOU, A. & in Potsdam, 6:50­77). Potsdam. P. SVENONIUS (eds.) Adverbs and Adjunction. Ｈｾｌｩｮｧｵｳｴ｣ HAJICOV A, E. (1995a) Postavenl rematizatoru v aktu:llnfm ｴｬ･ｮｾ vNy. Siovo a slovesnost 56:241­251. HAJICOVA, E. (1995b) Surface and Underlying Word Order. Travaux du Circle Linguistique de Prague. Prague Linguistic Circle Papers I: 113­124. HAJICOV A. E. (l999a) Aktu:llnl ｴｬ｣ｮｾｩ ｶｾｴｹ a yYstavba promluvy. In: HLADKA, Z. & P. KARLiK (eds.): Ccitina ­ Univeruilia a specijika 1. Sbomfk konfercnce ve Slapanicfch u Bma 17.­18.11.1998,47­54. Bmo. HAJIl:OvA, E. (l999b) Negation, Presupposition, and Focus. Paper presented at the Workshop on The Syntax and Semantics olSlavic Negation, Poznail. May 01­02,1999. HAJIl:OV A, E. (2000) Presupozice, alegace a akomodace. In: HLADKA, Z. & P. KARLiK (eds.) CcitinaUniverujlia a specijika 2. Sbomfk konference ve Slapanicfch u Bma 17.­19.11.1999, 123­130. Bmo. HETLAND, 1. (1993) Über Argumentstruktur. Fokus und Satzadverbiale. In: ROSENGREN, I. (ed.), Satz und Illokution 2, 109­126. TUbingen: Niemeyer. HOFFMANN, 1.M. (2000) A­Scrambling and Adverb­Placement. In: AIEXIADOU, A. & P. SVENONIUS (eds.): Adverbs and Adjunction. Ｈｾｌｩｮｧｵｳｴ｣ in Potsdam, 6:78­102). Potsdam. lACOBS, 1. (1991) Negation. In: v. STECHOW, A. & D. WUNDERLICH (eds.), Semantics. An International Handbook of Contemporary Research. '( ｾ Handbücher zur Sprach­ und Kommunikationswissenschaft, 6:560­596). BerlinINew York: Mouton de Gruyter. lUNGHANNS, U. (1997) Features and Movement. In: ALEXfADOU. A. et al. (eds.) ZAS Papers in Linguistics 9, 74­88. Berlin: Zentrum fur Allgemeine Sprachwissenschaft, Typologie und Universalien forschung. lUNGHANNS, U. (1999a) Zur VerbsteIlung im Deklarativsatz des Tschechischen. Ms., Universität Leipzig. JUNGHANNS, U. (l999b) Generative Beschreibung periphrastischer Konstruktionen des Tschechischen. In: ANSTATI, T. el al. (eds.) Linguistische Beiträge zur Slavistik aus Deutschland und Österreich. Vll. philologiae Slavicae, Sonderband Jungslavistlnnen­Treffen, TUbingenIBlaubeuren 1998. Ｈｾｓｰ･｣ｩｭｮ｡ 67: 133­165). MUnchen: Sagner. lUNGHANNS, U. & G. ZYBATOW (1997) Syntax and Information Structure of Russian Clauses. In: BROWNE, W. et al. (eds.) Annual Workshop on Formal Approaches to Slavic Linguistics. The Cornell Meeting 1995, 289­319. Ann Arbor: Michigan Slavic Publications. 616 FDSLIV JUNGHANNS, U. (in press) Scope Conllicts Involviug Sentential Negation in Czech. In: BROWN, S. & A. PRZEPIÖRKOWSKI (eds.) Negation in Slavic. B1oomington, Indiana: Slavica Publishers (Ms., 28 p.). KAYNE, R. S. (1995) The Antisymmerry olSyntax. Cambridge, MA: MIT Press. KOKTOvA, E. (l987) On the Scoping Properties of Negation, Focussing Particles and Sentence Adverbials. In: Theoreticallinguistics 14: 173­226. KOSTA, P. (1998) Über Argumentstruktur, Fokussierung und modale Satzadverbien im Tschechischen und Russischen. In: Zeitschrift fiJr Slawistik 43(2): 140­154. KOSTA, P. (2001) Negace a ｶｾｴｮａ struktura v ｣･ｾｴｩｮＮ In: KARLIK, P. & Z. HLAOKA (eds.) (2001): Cestina ­ Univerzalia a specijika 111, 117­138. Bmo. LANG, E. (1979) Zum Status der Satzadverbiale. In: Siovo a slovesnost XL.3:200­213. LANG, E. (1983) Einstellungsausdrücke und ausgedrückte Einstellungen. In: RÜLICKA, R. & W. MOTSCH (eds.), Untersuchungen zur Semantik. (= studia grammatica, 22:305­341). Berlin: Akademie. LANG, E., MA!ENBORN, C. & C. FABRICIUS­HANSEN (eds.) (2003) Modifying Adjuncts. Berlin, New York: Mouton de Gruyter. MAY, R. (l985) Logical Form: Its Structure and Derivation. Ｈｾ Linguistic Inquiry Monograplts, 12). Cambridge, MAILondon, England: MIT Press. [Me] (1987) = Mluvnice teStiny. Eds. J. Pell' et al. Praha: Academia, vol. 3: skJadba. PILl, D. (2003) On A­ anti A '­Dislocation in tlte Lejr Periphery: A Comparative Approach to the Cartography olthe CP­System. (= Linguistics in Potselam, 20). Potsdam. POLLOCK, J.­Y. (1989) Verb Movement, Universal Grammar, and the StructW'e of IP. In: Linguistic Inquiry 20(3):365­424. PROGOVAC, LJ. (1998) 'Avoid Conjunction', Adjunction, and the 'Coordination of Likes Constraint'. In: Ｌｃｉｖｏｋｾ ｂ 2., S. FRANKS & W. SNYOER (eds.) Annual Worlcshop on Formal Approacltes to Slavic Linguistics. The Connecticut Meeting 1997, 252­266. Ann Arbor: Michigan Slavic Publications. RECHZIEGLOVA, A. (1995): On Negation in Czech. Praha. R.Izz1, L. (1997) The Fine Structure ofthe Left Periphery. In: HAEGEMAN, L. (ed.) Elements olGrammar. (= Kluwer International Handbook., 01 Linguistics, 1:281­33 7). Donirecht: Kluwer: SGALL, P. et al. (l980) AktualnE tlenenE vely v ceStine. (= Studie a prace lingvisticke, 12). Praha: Academia SGALL. P. et al. (1986) The Meaning olthe Sentence in Ils Semantic and Pragmatic Aspects. Dordrecht: Reidel. SHAER, B. (2000) Syntactic Position and the Readings of 'Manner' Adverbs. In: FABRICIUS­HANSEN, c., E. LANG & C. MAIENBORN (eds.) Approaching the Grammar 01 Adjuncts. Proceedings of the Oslo Conference, September 22­25, 1999. (= ZAS Papers in Linguistics, 17:265­286). STEINITZ, R. (1969) Adverbial­Syntax. (= studia grammatica, 10). Berlin: Akademie­Verlag. UHL!{(OVA, L. K postavenl tzv. ｨ｣ｹｮｴｾｶ pftslovcl v aktualnlm ｣ｬ･ｮｾ ｬＮ Siovo a slovesnost LX(2): 143­148. VESELOvsKA, L. (1995) Phrasal Movement and X'­Morphology. Ward Order Para/leis in Czech and English Nominal anti Verbal Projections. Doctoral dissertation, Univerzita Palackeho, Olomouc. WYNER, A. Z. (l998) A DiscoW'se Theory ofManner and Factive Adverbial Modification. In: HALST/IN, J. & A. NUHOLT (eds.) TWLTB Formal Semantics anti Pragmatics 01 Dialogue, Enschede, Universiteit Twente (Ms.). ZUBIZARRETfA, ML (1998) Prosod}', Focus and Word­Order. Cambridge, MA: MIT Press. Peler Kosta Department of Slavistics University of Potsdam P.O. Box 60 1553 D­I44 I5 Potsdam Gennany e­mail: [email protected]­potsdam.de Formal Slavic Linguistics is concerned with explicit description of prosody, phonology, morphology, syntax, semantics, pragmatics, information structure and language acquisition or impairments of language (aphasia) of Slavic languages within a certain theoretical framework of Principles and Parameters (Chomsky 1995 passim). But the two parts also iIIustrate the diversity of approaches we use in attempting to reflect the entire range of suhfields within a given theoretical framework of cognitive science. Peter Kosta is Professor of Westslavic Linguistics and Chair at the Slavic Departrnent at Potsdam University. Joanna Btaszczak is Psychological and Technical Assistant of the Theory of Gramrnar (Syntax and Morphology) at the General Linguistics Department at Potsdam University. Jens Frasek is Assistant of Westslavic Linguistics at the Slavic Department at Potsdam University. Ljudmila Geist is PhD student at the University in Berlin and employed as a research assistant in the DFG­sponsored project Semantic Interfaces: Copula­predicative constructions at the Centre for General Linguistics (ZAS) in Berlin. Marzena Zygis is research assistant at the Centre for General Linguistics (ZAS) in Berlin and involved in the project The phonological word. In 2001 the editors organized the Fourth European Conference on Formal Description of Slavic Languages (FDSL IV) in Potsdam. The FDSL­conferences take place biannually in Leipzig and Potsdam. www.peterlang.de