Nivkh Time Expressions (1)

1.  Introduction
 All Languages have some types of time expressions. But it is not the case that every language has the same set of the types. It has its own particular set of time expressions. However, most languages share the following types:

(1)  Time Expressions
   a.  Time Indicators (abridged: TI) which represent time units of daily life or some
intervals, e.g. today, soon, etc.
   b.  The socalled Tense which overtly marks when a verbally represented event
   occurs relative to
the time point of speech. It can also mark the relative order
    of events each other

   c.  The socalled Aspect which thematizes a certain phase within the process of a
represented event. It revises the inherent process a verb represents,
   namely LA in (d) and
makes up a new complex of the process. 
   d.  The socalled aktionsarten or, in my terminology, Lexical Aspect (abridged: LA)
   of verbal
categories which represents how an event is going. Each language has
   its own particular set
of types of LA.

Time Indicators like now, today, winter, etc. can be found in most languages. A language without any time indicator like today is hardly conceivable. But many languages lack, e.g. the word for winter, autumn etc., because they need not such words from the local and meteological ground. Therefore, the possession of TI in general belongs to the universal features of languages, but what kinds of TI a language has belongs to the individual character of a particular language.
 Tense is not an indispensable grammatical category of languages. It is only an optional category which many languages really lack. Therefore, languages can be typologically classified by means of the feature whether they possess Tense. Such a classification would bring about a simple binary opposition: tense-languages versus tenseless-languages. But the thing is more complicated. Some languages represent it morphologically by means of affixation, the other by some complex syntactic device. And some others distinguish the relative order of events with an adverb. If a language marks Tense obligatorily with a certain adverb, we can hardly classify the language into the tenseless group, for it can be estimated as grammaticalized in a certain degree. But if the marking occurs only optionally, we used to classify the language into the tenseless type. The adverbially tense-marked languages are, therefore, classified further into the subclasses: obligatorily or optionally marked types. So, we have a variety of different marking of Tense, e.g. the morphological, syntactic and lexical (adverbial) marking. This formal types of marking is cross-classified with the semantic variety of marking: the distinction of past marking, future marking, or aorist, etc. and the optionality of making with various grade of its grammaticalization, etc. Therefore, we have to establish a multiple typology of tense forms and their functions, so that languages are classified by the criterions, first whether they possess it, second, what they marks and how. Tense belongs anyway typically to typological categories.

 The socalled Aspect is a typological category, too. But it is more difficult to find languages without aspect than tenseless ones. In the history of many languages, aspect precedes tense and the latter has been derived from the former. At least from the historical point of view, Aspect seems to be more underlying than Tense. But from the semantic viewpoint, Aspect stands on a different dimension than Tense. Tense is concerned with the historical order of events relative to the time point of speech, but Aspect is not. It is confined within the process of an event and determines how an event occurs and is going on, namely, it begins, continues, finishes, and so on, but it is not concerned within other events which a speech mentions eventually. In some languages, the grammatical forms for Aspect are derived from verbs, but in some others it is represented with affixes attached to verbs. And there are languages, too, which represent both categories with the same form under different syntactic conditions, e.g. Mandarin Chinese le. Aspect is not necessarily supported by formal devices, it is definable merely by means of its semantic functions.

 Languages represent things and events. Suppose that every language has verbs, then this category represents events, though the contrary is not true, for events are not always represented by verbs. Suppose again that the meaning of verbs corresponds to the verbally represented events substantially, then verbs have to represent various types of events. They represent events like a human act, a natural phenomenon, or the state and/or quality of things etc. So, the verb meanings can be classified according to the sorts of events so far as they are reflected in them in substance. We call these feature of verb meanings the aktionsarten or LA of verbs. However, most languages do not formally mark these types of the semantic classification of verbs in grammatical forms. But this does not say that languages have absolutely no formal support to indicate LA. On the contrary, in many languages, the differences of LA are indicated with affixes, and in the other it is implied indirectly by some morphological and syntactic restrictions on the possible concatenation of morphemes. In such cases, we can describe the LA by means of the morphosyntactic conditions.
 Time expressions of languages represent events. Time Indicators (1a) indicates events themselves or when or how they occur. Tense (1b) combines events with the time point of speech and events each other. Aspect (1c) thematizes a certain phase of an event, and aktionsarten or LAs (1d) represent the processual types of events, if necessarily, with some other semantic feathers altogether. For the appropriate description of time expressions, therefore, we have to show first what types of events do occur and how do they occur, too. For that purpose, it is reasonable to make use of a strong logical model for event types. We utilize the following model which was first developed in Stuttgart 1970's. 

(2) Model for LA and Aspect
   a. An event occurs at a time t1 and continues for a time being T and ends at
   a time t2.

   b. An event presupposes a situation before the occurrence of it and entails another
   situation after
 the end of it.
   c. An event is, therefore, the chain of three situations: presupposed situation
   (presit)- verbally
represented situation (verbsit, abridged as sit) - entailed situation

   d. Every situation touches densely with one another either with an open (*) or with
   a closed time
 interval (#). The closed interval belongs to the verbally represented
    situation and the open
interval to the other situation.
   e: t1,t2,T (t1$T$t2) [ presit$(T) sit $ postsit ]
     where $ = {, , , , }
   f: There are times t1, t2, T, such that: t1$T$t2 , for them the relation is applied:
    [ presit $ sit occurs at T $ postsit ]

 The relative order of events is also described by means of the following model:

(3) Model for Tense
   Two events e1 and e2 are ordered such that
   t1,t2 [(t1) e1 $(t2) e2 ]    
   where $ = {, , , ,

2.  Time Expressions of the Language Nivkh
2.1. Types of Time Expression of Nivkh; an Outline
 The time expressions of the language Nivkh have been hitherto investigated merely partially. We have some partial grammatical descriptions on verbal aspects and tense. Needless to say, such questions like what kinds of types Nivkh verbs have and how their types interact each other have not been even asked. We have to start, therefore, from getting an image what types of time expressions the language Nivkh have. We will first collect some of time indicators from texts and field datas, and the grammatical devices for time indications from the grammatical descriptions hitherto made by Russian and Japanese grammarians in the last century. About tense and aspect of verbal expressions, we have relatively sufficient data to begin discussion. But about the lexical aspect of verbs we have only scanty descriptions. Panfilov 1965 mentioned about protekaniya deistvia (aktionsarten) of verb, but he seems to understand this notion not as an independent grammatical category, but he does not clearly distinguish it from morphological aspects. It was Galina Oaina who atempted to investigate this category for the first time. However, she restricted her grammatical description only to the socalled quality verbs (kacestvennye glagoly). Had she been able to work longer, she could have surely contributed more than her teacher (Panfilov) to her own native language.  Now, we will first sketch the types of time expressions of Nivhk.

I. Time Indicators
 Nivkh has many types of Time Indicators which represent when and how a event occurs, e.g.

(4) Time Indicators of Nivkh
   a.  [time adverbs]: p«t (tomorrow), naf (now) , tJ«k (long tome)
   b.  [nouns]: tH«t (morning), parv (evening), urk (night), ku (day and night)
   c.  [verb phrases]: nana ozdJ (just getting up), anJ malGodJ (many years)
   d. [compound words]: tH«t-tH«t (every day), l«z anJ (last year), tHof tolf
(last summer),  nana ozdJ ( just after getting up)    
      (from Puxta 1995/2003 (PU in the following), III-17, 4001- 4061)

II. Tense forms
 Nivkh has been hitherto thought to have two Tense forms: non-marked finite form (aorist), i.e. -φ and the future form with the suffix -n«- which indicates the future occurrence of an event. Without this morpheme, verb forms are used for all the other occurrences of events, namely, for events in present as well as in past time. But the unmarked form can also represent a future event, if it is a matter of course that the event occurs in the future. The suffix -n«- is used, namely, mainly in a marked case if an event prescribed to occur in a future time. Compare the following sentences:

(5) Examples for Tense
    a.  nJnanak              pH!al dJu--dJ-ra,              ni pH!al latJsku guku-n«-dJ-ra.
         my-elder sister     floor wet-FUT-FIN-AF    I floor  mat  clean-FUT-FIN-AF
       My elder sister wet the floor and I clean the floor mat. 
   b.  pitGu     urudJ  smodJÄu   bak   pHr«u  urdJÄura. (PU516)
       book-PL well  love-PL  only  learn  well-PL-AF
       Whoever loves books well, they only learn well.

III Aspect forms 
Panfilov 1965 (PA65,
§38 -§44) distinguishes the following 4 types of morphological aspect forms:

 (6) Aspect forms by Panfilov
    (i)  -Ä«t-/-k«r- /-Ä«r- for conclusive, perfect or resultative aspect
   (abridged: CON)
 (Amur-Dialect, its corresponding Sakhalin-Dialect form:
Äar-/-xar-) , e.g.
    a. caj    «rk         muz -Ä«t-tJ .(PA70)
      Tea  already  become=cool-CON-FIN
      Tea is already become cool.
    b. nJ«N  «rHt««rHt« modJra:  c«N «Är«kon         mu-Ä«t-tJ.
      we  always   grieved: you long-time-ago die-CON-FIN
      We always grieved that you had died long time ago.
    (ii) -ivu-/-ivi- for stative or durative aspect (abridged:DUR), e.g.
    c. coN«N«r virot         эÄr    pHr«-ivu-dJ . (PA76)
      fishing  go-PART soon  come-back-DUR-FIN
      Having gone fishing, (he) was soon coming back.
     (iii)  affixation of -tJu- or duplication of verb stem for iterative aspect
     (abridged: ITER),e.g.

    d. jurudJ (read) vs. juru-tHu-dJ (read many times)
       эzqou-dJ (wash something) vs. cXou-tHu-dJ (do washing) (PA73)
    e. h«   macala       N«cki          NaqrNaqr    tHm«Nan            h« qan N
NN«N      somru

     this young-man mountain-top one-by-one step-over-PART this dog
    one-by-one let-off

          huNdJ-huNdJ-dJ. (PA73)
      The young man, stepping over the mountain tops one-by-one, left a dog
    behind one-by-one.

  (iv) affixation of -x«, duplication of verb stem, adverbal participle ending
   -t, -r and syntactic

    concatenation with hadH (exist) for conventional/habitual aspect
  (abridged: HAB), e.g.

    f.  h«  nivx qHotr  l«u-x«-dH. (PA75)
       this man bear  hunt-HAB-FIN
     He hunts bears.
  g.  paluin   has   vxtJ-vaxtJ  эGdH. (PA75)
     forest-in cloth tear-tear  quick
     In the forest clothes are raged quickly.
  h.  nJ«Kn  t«fku        tHirkir      l«tHadJ-Äu. (PA75)
     we-by  house-PL tree-with  make-exist-FIN-PL  (l«tHadJ l«ttH + hadJ )
     By us houses are made with log.

 As for the function of aspectual morphems, Gruzdeva 1998 (Ch.3.4.1.) mentioned that "aspect markers transform the original meaning of verbs", in the sense, that they play an additional semantic role to change the inherent lexical meaning of verbs into a complex time representation. To take an example of the morpheme -x«, the event represented by the inherent meaning of a verb Vi is changed into a conventional/habitual process, i.e. the event of the Vi into a conventional/habitual occurrence. Here remain yet some crucial problems, it is not clear how the inherent meaning of verbs "interacts" (in the sense of Gruzdeva) with aspectual markers, what conditions underlie the interaction, how they are related with other aspectual complexes and other relevant questions. In order to answer these questions, we have first to extract the relevant part of the inherent lexical meanings of verbs which goes into the interation with aspect forms, and distinguish it from the substantially cognitive part of them which does not go into the merging process. Let us call the former the processual meaning and the latter the cognitve meaning of verbs. The former is LA of verbs and the latter the individual meaning of them. The distinction is well motivated because the former parts make types and the latter, on the contrary, language-particular lexical notions. However, we do not yet know exactly how many and what kinds of types of the inherent processual meaning Nivkh verbs have and how they interact with what aspect forms. We have to build up descriptions little by little. (Cf.: Kaneko 1995, new edition 2003 (in CD-form available) as a descriptive model)

IV Lexical Aspect of Verbs
 The typology of LA of Nivkh verbs has not yet been seriously investigated. Panfilov 1965 mentiones the lexical "protekanie dejstvia" of verbs in many places, especially in relation to its possible change when aspect forms are attached to them. He seems to understand "aspect" (vid) as the complex notions as follows (#39ff.):

(7) Panfilov's Notion of Aspect
   a. Aspect is the represenation of the temporal and local features of process
dejstvia") which verbs express as their inherent meaning.
   b. Aspect is "the grammatical category of non-syntactic word change". It is
as the opposition of lexical units or verbal forms which are changed
   by means of some
 morphological procedures.
   c. Nivkh has 4 aspect forms: conclusive, iterative, conventional/habitual or
   durative aspect. 

   d. Almost all verbs are aspectually neutral without any (morphological) aspect

   e. But there are some pairs of verbs which have inherently with an aspectual
   opposition, e.g.

      qmadJ (run) versus kHimmJ (run away), janH /jajmadJ (look at) versus ind«dJ (see).

  As far as we can judge from (7) above, Panfilov 1965 understands the notion of aspect as such possible semantic oppositions in a certain temporal/processual meaning which are found between cognitively synonymous pairs of verbs. He does not assume that the inherent meaning of a word necessarily contains a certain temporal feature which determines the processual type of events it represents, even in case it does not make any aspectual opposition with another verb. In other words, he does not think of LA, i.e. an aktionsart in the inherent meaning of verbs. His understanding of "aspect (vid)" has, therefore, twofold fatal shortcomings: first, he overlooked the fact that every verb represents its own processual/temporal (eventually local) feature ("protekanie dejstvia") of events (see (7a)). Second, he trivializes the grammatical category of aspect into the lexical opposition among cognitively synonymous verb pairs. He assumes that the aspectual meaning of verbs can be represented only if the aspectual opposition is actualized overtly as a lexical pair. This idea of him goes back to the traditional linguistic description of Slavic grammarians who assert that the lexical meaning of (some classes of) verbs make an aspectual opposition (like perfective versus imperfective) within the lexical system of a (slavic) language. He was, too, infected by the "national" prejudice to look at Paleoasiatic languages from the traditional Slavic viewpoint.

2.2. Quality Verbs of Otaina (1978)
 Galina Otaina attempted to show the semantic classification of Nivkh verbs. Cleary, she did not intended to mention the whole classes of Nivkh verbs, but confined only to the lexico-semantic class of verbs which she calls quality verbs (kacestvennye glagoly). She distinguished them from all the other non-quality ones. The latter classes is practically a gathering of various verbs which do not represent quality. She thinks of the following types of non-quality verbs (GO13):

(8) Otaina's Types of Quality Verbs
   a. verbs representing various actions, e.g. nyzdJ (throw), jazdJ (bite),
   b. verbs representing states, e.g. qHodJ (sleep), qodJ (hurt), hurtHivdJ (sit),
  c. verbs representing natural phenomena, e.g. tidJ (thunder), kydJ
  (fall (rain/snow)), qHadJ (retreat (water)).

 All the other types of verbs are, according to her, the class of her quality verbs, which has three main semantic subclasses (GO 17-22):

(9) Otaina's Classification of Quality Verbs
   a. expressions for human features with two subgroups
     a1: terms for physical features, e.g. onJvildJ (tall), h«jmdJ (old),  pizvurdJ (pretty)
     a2: terms for modal and mental feartures, e.g. qHyzNardJ (happy), urdJ (good);
         qannaldJ (unscrupulous), osqavildJ (coward)
   b. expressions for material features with four subgroups
     b1. color terms, e.g. v«lv«l«dJ (black), qHonudJ (white),  paÄladJ (red)
     b2. terms for size and forms, e.g. verdJ (broad), NaladJ (equal), uldJ (high)
     b3. other physical feature of things,e.g. qHavudJ (hot), meqadJ (strong),  naudJ (sufficient)
     b4. terms for quality of foods, e.g. nenidJ (sweet),  hordJ (tasty), perdJ (salty)
   c. expressions for various features of animals with two subgroups
     c1. color terms for fur, e.g. pa:rdJ (red),  «:rdJ (black)
     c2. other physical terms, e.g. NardJ (fat), vesqardJ (powerful)

 She gives a detailed description about the syntactic behavior of the quality verbs. Some of the syntactic features she described are useful to distinguish them from the non-quality verbs. We will here only summarize the relevant features of quality verbs:

(10) Morphosyntactic Features of Quality Verbs of Otaina
    a. Quality verbs have no imperative mood. They can combine neither with -ja
    for 2nd person
singular, nor -ee for 2nd person plural.(GO83/84)
       She mentions two exceptional usages:
     (1) In some prohibitional usage, they can be affixed with -ja and -ee,
        e.g. tHa h«m-ja (don't be old, tHa: morpheme for prohibition ),
tHa um-ja (don't be angry)                     
     (2) The optative usage in 3rd person is possible:
        e.g. kHe pil-ÄaZo (The net may be big, ÄaZo: optative suffix)
    b. Non-quality verbs in future form with suffix -n« have always "purely temporal
but quality verbs are ambiguous, in the sense that they can have
   temporal and modal
meaning, (GO67-69),
       e.g. for non-quality verbs: «m«k kHэZ-n«-dJ (Father will set nets.)
          for quality verbs: t«f pil-n«-dJ : ((1) The house is going to be big.
                                              (2) The house may be big.)  

 She mentions the aspectual behavior of quality verbs (#45-52) as follows,
(a) With the aspect form -Ä«ta, they mean that the inherent meaning is preserved.
  She calls this
aspect sokhranenie dejstivija (preservation of situation).
  We mark this meaning with PSV.

(b) With the auxiliary verb hadJ, they mean a conventional/habitual occurrence of
 event. (obychnost'
dejstvija) Note that the conventional/habitual aspect form -x
cannot be attached to quality verbs.  We mark this meaning with HAB.
(c) With the durative aspect form have -ivu/-ivi, they mean that the state comes
  about. (stanovlenie
 dejstvija). We mark this meaning with CAB(=COME-ABOUT)

To illustrate these combinations, she shows the following examples:

(11) Otaina's Aspect Types of Quality Verbs
    a. preservation of quality:
      co  loqr -Ä«ta     -dJ
      fish frish-PRV-FIN
      Fishes remain frish.
    b. conventional/habitual events
      t«  «rux    l«x       «:rti     «SkirHa-dJ («Skia-dJ=«kiradJ+hadJ)
      this season weather always bad-HAB-FIN
      In this season, the weather is always bad.
     NB. She remarks that the iterative form with -x« is not possible for quality
    verbs and
emphasizes that quality verbs are only possible to represent
    habitual acts with
hadJ and in the adverbial forms -t/-r, e.g. Narked by EMPH.  
        tilf         caX   m«rke veurHadJ  (veu-r+hadJ )
        autumn water high   deep-EMPH-FIN
        In autunm water will be high and deep.
    c. coming about of quality
     l«Äigma        «rk      tarZe-jvu       -dJ
     dried salmon already dry-CAB-FIN
     Salmon is already becoming dry.

 Note that she asserts that quality verbs have different sort of combinational aspects from non-quality verbs. The difference lies in four points:

(i) They have neither conclusive/resultative (zakoncennyj) nor iterative
 (mnogokratkij) aspect,
 instead they have preservative (sokhranenie) and
  inchoative (stanovlenie) aspect.

(ii) They make no conventional/habitual aspect with the suffix -x«, but they
   are possible to represent
this aspect meaning with a construction:
  [adverbial form -t/-r, -
Nan + hadJ ].
(iii) They make preservative form when they are combined with the aspectual
  form -
(iv) The suffix -iv(i)/-iv(u) makes an inchoative aspect, though the same form
   makes a durative
  aspect with non-quality verbs.

In order to examine the interrelation of aspect forms and different verb classes, let us compare their different behavior of quality and non-quality verbs when they are connected with different aspect forms:

 (12) Aspect Markers of Quality and Non-quality Verbs

   aspect markers

   quality verbs

   non-quality verbs


















  -tHu **















   *: Panfilov 1965 and ** :Otaina 1975 do not mention these forms, respectively.

 Let us pick up some relevant points in the table. First, Otaina did not mention Panfilov's aspect marker -Ä«t. She dealt with the suffix -Ä«ta exclusively, even without saying if this suffix has something to do with the conclusive -Ä«t. Interesting is that Panfilov 1965, on the other hand, did not mention Otiana's marker-Ä«ta, too. It is more interesting to find two examples he cited (PA72) which show quality verbs are suffixed by -Ä«t (PA72).

(13)  -Ä«t to Quality Verbs (Panfilov's examples)
     a. TuZ-Ä«t          caj  amra  qavrdJ.
       cold-become tea taste  bad(?) [qavrdJ = «kidJ ?]
       Cool tea has not good taste.
    b. ur-Ä«t-tJ
       good-become [composita]

So far, it is not yet clear to me whether Otaina's -Ä«ta has something to do with Panfilov's -Ä«t.  At the moment, we let the problem open.
 Second, Otaina's aspect markers -iv(i)/-iv(u) is mentioned by Panfilov 1965, too. But he says that it has different aspectual meaning, namely DURativation. This implies that this form brings about different meanings according as it are attached to different classes of verbs, quality and non-quality verbs. Here we have practically two questions. (i) We have to ask whether this aspect marker has a constant function, no matter if it is suffixed to different classes of verbs. If yes, the difference of aspect meaning is not due to the function of the suffixes, but it has to be ascribed to the meaning of the verb classes; if it is affixed to non-quality verbs, it represenst CONclusive aspect, but if to non-quality verbs, ComingABout of the quality. Therefore, we have to ascribe this meaning difference to the interaction of the aspect marker and the different semantic feature of verb classes. I like this view and assume that the aspect form has only one meaning, which realizes the different meaning according to the verb class which precede it. Instead, if we do not stand on this viepoint, we will go wrong into a "theory" that aspect markers themvelves had varisous meaning, i.e. that they are  ambiguous or homonymous, just like some Japanese grammarian who assert that -te iru is homonymous.
(ii) The second question is concerned with the morphological status of the aspect markers. We will formulate it simply: are they suffixes or verb stems? For example, if (13b) ur-Ä«t-tJ  can be analyzed as [quality verb + aspect form + fnite suffix], does the finite suffix -tJ  belong to quality verbs discontinuously or to the aspect form? If it belongs to the aspect form, it has to be interpreted as verb, on the contrary, if to the quality verb, the aspect form -Ä«t only a suffix without verbal character. In general, if aspect forms are suffixes, the marker -dJ belongs to the beforegoing verb stems. The difference can be illustrated as follows:

 (13) Morphological Status of Aspect Marker
    a. [V1(stem) + [V2(aspect marker) + -dJ (finite marker)] V] V = aspectual
   complex verb

    b. [V1(stem) + [SUF(aspect marker)] + -dJ (finite marker) ] V= aspectually
   suffixed verb

If (13a) is a correct analysis, aspects markers are themselves verb stems. Then, Nivkh has the concatenation of verb stems to construct verb complexes which ends with a finite marker at the end of them, just like in Japanese -rare, -sase and some other forms. We find also more analogies in this connection: the markers for conventional/habitual aspects -x«, -t/-r and +hadJ (11-5) above are parallel in a morphological view point to Japanese stem-concatenations, te-forms and advervial form (ren-yookei) + independent verbs. This problem of the morphematic status of -dJ  is also related to an another relevant issue about the morpheme concatenation which Krejnovich 1958 mentioned. The finite marker can be exchanged with a noun to make a free, i.e. non-bound, form in itself, which can make with other forms further compound nominals. Here, we find again a parallel phenomenon both in nominal and verbal spheres. The analysis (13a) is, therefore, well motivated. Namely, if this is true, we can conclude that the aspect markers are verbs themselves, just like in Japanese.
 Third, it is not difficult to imagine that quality verbs have no iterative aspect in straight-forward way. And the conventional/habitual aspect is restricted to reduplication and the verb complex with +hadJ , but the simple suffixation of -x« is excluded. The description of Panfilov 1965 appears to regard this morpheme as the inherent means to make a conventional/habitual aspect. The examples he cites are sentences with non-quality verbs like va-x«-dJ (fight(dogs)), l«t-x«-(making), vivi-x«-dJ (walk/go), which represent animate actions. This morpheme is perhaps the formal means to represent a habitual action. The dictionaries now available contain neither this particle nor its possible derivation. It needs a further investigation. 

2.3. Lexical Aspect of Nivkh Quality Verbs
 Otaina's analysis of quality verbs has brought about the crucial problems for lexical aspect of Nivkh verbs. As the table (12) shows, the aspect marker -iv(i)/-iv(u) make different aspectual meanings according as they are suffixed to quality and non-quality verbs. We ascribe this meaning difference not to the possible homonymity of the aspect marker, but to the concatenation type, i.e. to the inherent meaning difference of the verbs which precede the aspect marker.
 Now, if we presuppose that the meaning of the aspect marker is constant, we have to ask first what it does mean. Look at some relevant examples from Panfilov 1965 (§40 and§43) in case of the concatenation with non-quality verbs:

(14) Non-quality Verbs + -iv(i)/-iv(u) (Panfilov's examples)
    a. caj  qHavu-ivi   -dJ    (vs. caj q'avudJ )
     tea  boil  -DUR-FIN
     Tea  is boiling.        (vs. "tea boiled")
    b. hoqot эxt   vidJ-Äu.   imN matqamatqa dJi:r v«jrox    vit            
         then   home go-FIN. they thick-thick   tree under-to go-t(PART)      qHojn«-dJ-Äu.

      Then they went home. They, going under a thick tree, slept together.

        NaN«N-NivÄ-gu     pHr«         -ivi-dJ-Äu     imN   m«-dJ-Äu.
       hunter-man-PL   come back -DUR -PL  they  hear-FIN-PL.
       The hunters are coming back, they heard.

    c. Ni     mu   l«t-ivu-       nJivx    nrH«-dJ.
      1PS ship  make-DUR man   look-FIN
      I saw a ship making man.

    d. pila  nJivx  m«Ä-rH           por-dJ.    tHuÄř  tHoZ   -Ä«t  -ivi  
    old  man  come-down-PART lie-FIN.  fire   go-out-
-dJ    ø«-  ha -dJ.
CON-DUR-FIN warm-be-FIN.

        The old man came down and lay.   Fire  has been going out and was warm.

 From these examples, it is obvious that the aspect marker -ivi/ivu indicates the going progress of an action (14c), a movement (14b) or a process (14a). The example (14d) shows that the marker can be combined with another aspect marker -Ä«t  (for conclusive/resultative aspect) and represents the resultative state after the process of the beforegoing verb finished, namely the resultative state in progress.
 Panfilov 1965 does not mention the concatenation with quality verbs. But he notices that punctual verbs like эsptJ (sting, stick) and the inherent durative verbs like kH«m«l«dJ (think) can not be suffixed by this aspect marker.
 As far as we have seen in the examples above, we can conclude that the aspect marker -ivi/ivu indicates no more than the durative progress of a process represented by the preceding verb. Now, we compare some relevant examples of quality verbs (GO 1978) with Panfilov's examples above.

(15) Quality Verbs +-iv(i)/-iv(u) (Otaina's examples)
    a. mer эri pHakefake              N«Z     -jvu    -dJ.
      our river more-and-more(?) shallow-INCH-FIN
      Our river became shallower and shallower.
    b. if  pH-nanak«k      tob-jvu-dJ.
      he self-elder-sister old-INCH-FIN
      He became elder than his sister.

   c. nin nin pHu     -t       vi  -ke,                 mer d«f     hoNÄa-jvu   -dJ.
      one one go-out-PART go-CONJ(=after), our house vacant-INCH-FIN.
      After (they) have gone out one by one, our house became vacent.

   d. if  pil-jvo        comsomy-jvu-dJ.
      he big-CONJ(then), quiet-INCH-FIN.
      If he grows, (then) he will become quiete.

 The examples (15a and b) are straight-forward. As the finite marker -dJ  markes the time preceding the speech time, the quality represented by the verbs have already come into existence when the sentence uttered. This time relation corresponds to Otaina's interpretation that this aspect form marks the time of coming about of the quality. In the examples (15c and d), the adverbial phrases preceding the aspect forms (marked with CONJ) indicate the time when the quality comes about. In (15d), the time is not set in aorist, because the adverbial phrase indicates the time when the condition is satisfied, namely, when the mentioned time (when he grows up) comes about. In any case, Otaina's interpretion is justifed.
 Now, assume that the time indicated by -dJ  or the adverbial phrase in the sentences (15) is symbolized by a certain time interval t, we can say that the quality Q is established in the time t. In the sentences (15a,b and c), the time t precedes the speech time t0 because of the finite marker
- dJ. But in (15d), the relative order of the time t and the speech time t0 is in reverse order because of the conditional clause which indicates the time when the situation Q comes about.

(16) Coming About (stanovlenie) of the Situation
    a. t (tt0) (t) Q for (15a,b and c)
    b. t (t0t) (t) Q for (15d, because of the subordinate clause)
     where t0: speech time

 The formulation (16) says that in unmarked cases like (16a), the quality Q comes about at t which precedes the speech point.
  Now, we will think about again what quality verbs mean in a non-aspectual usage. First, we will look back the finite sentences with socalled aorist tense-dJ, e.g.

(17) Non-quality Verbs with -dJ
    a. n«x kHr«u-dJ эrq ur-dJ. (GO22)
     today rest    ?  good-FIN (эrq эrqtJpress?, Otaina's Russian
     Today may be good to rest.
  b. if «rk ur-dJ.
      he already good-FIN
      He is already well.
    c. t«   эri     co ram-dJ. (GO32)
      this river fish many-FIN
      This river has many fish.    
       d.«Ärkon/tHof  t«     эri   co  ram-dJ. (test sentence)
      formerly this river fish many-FIN
      Formerly, this river had many fish.

The sentences (17b and c) represent the situations "to be well" and "to be many", respectively. These situations include the speech time and the indefinite time interval preceding the speech time. This case is regarded as the normal represetation of aorist. The sentence (17a) includes the time adverb "today", so that the represented situation includes the speech time. The modal word эrq (Russian by) shows the speaker's feeling that the present situation goes on for a time being. The test sentence (17d) contains the time adverb «Ärkon/tHof  (some time ago/in the old time), so that it can imply that fishes are now not so many as before, i.e. the described situation is not true at the speech time. My indigenous friend (Galina Demjanovna Lok) told me that the sentence is good in this meaning. So, we assume that the aorist can indicate the time interval which precedes the speech time. Therefore, we can formulate the aorist time relation as follows:

(18) Aorist Time of Non-quality Verbs
    t (tt0/tt0) (t) Q (or abridged: t (tt0) (t) Q)

  As Otaina 1978 asserts, the future tensed sentence of quality verbs are ambiguous: they have a future or a modal meaning. Looking back at her examples (10b) and some more examples:

(19) Non-quality Verbs in Future Tense
    a. if   pH«m«k      øalagur     kH«   -n«   -dJ.(GO68)
     she own-mother resembling skillful-FUT-FIN
     She will be skillful like her mother.
    b.  t«    qHotrnonq-xu «rk       эgut pil-n«-dJ.(ibid.)
        this bear-babies   already rapid big-FUT-FIN
        These bear-babies will be already rapidly big.
    c.  t« anJ velso      pav      -n«   -dJ   -ra.(ibid.)
       this year salmon prohibit-FUT-FIN-EMPH
       This year salmons are prohibited to take.
    d.  «tik             pHi  Äe  bos   nok-n«-dJ          xe-dJ.(ibid.)
       grandmother self buy cloth narrow-FUT-FIN think-FIN
       Grandmother thought that cloth she for her self bought would be narrow.

 The sentences (19a,b and c) show that the qualities concerned will come about in the time indicated or implied by the time indicators. The complex sentence (19d) is an interesting example. Here, the future form of quality verb  nok-n«-dJ is subordinated by the matrix verb xe-dJ  (think/thought). According to Otaina (GO68), the embedded quality verb means that the quality comes about after the time represented by the matrix verb. Namely, the time when the cloth is narrow stands after or at least equal to the time she thought because of the future morpheme -n« : xe-dJ t0 nok-n«-dJ .   
 As the modal usage of -n« indicates, needless to say, that the quality represented by the verb will come about after the speech time, quality verbs with the future suffix represent in any case that the quality comes about in the time after the speech time. Therefore, it is true that

(20) Time of Quality Verbs with -n«
     t (t0t) (t) Q

 Suppose that the lexcial aspect of quality verbs Q is formulated as "$…DUR…$", because, first, it has a long situation, therefore "…DUR…" (We markes the internal duration of LA as small "DUR", too. The marking corresponds to the aspectual duration DUR because we assume that the latter is derived from DUR of LA). Second, we leave so far the beginning and the final frame not yet defined and write it as $…$. Third, as the situation comes about at a time t, the situation of quality verbs Q is surrounded by the speech time t0.

(21) Time Relation of Nivkh Quality Verbs

a.  concatenation types

b.   time relations

c. t0-position around situation Q

I. with aspect form -ivu/-ivi

  t (t?t0)(t) Q

   t0 ?  Q$DUR$

II. in finite form -dJ

    t(t) Q


III. in future form with -n«

 t (t0t)(t) Q


 Now, we have to ask how it looks like such a situation which comes about satisfying the conditions in (21). It has to be able to come about before as well as after the speech time, and at the same time the speech time can be included in it. It has to be nothing than an event whose situation is open both to past and to present. Such a situation can be formulated as

(22) LA of Nivkh Quality Verbs
    Q = situation of the form: *…DUR…*

 Otaina's quality verbs represent such an event whose processual feature is characterized as the situation of this type. The table (21) shows that it realizes the different time meanings according as it is combined with three different morphemes in (21a,b and c). We conclude that the language Nivkh has verbs with the situation type *…DUR…*, just as Japanese, but they have the language particular morphosyntactic behavior as the table shows.

2.4. Lexical Aspect of Non-quality Verbs
 Now, let us go back to the Panfilov's examples (14) which all contain the aspect marker -iv(i)/-iv(u). We pick up the verb forms in order to see how they are concatenated and what temporal meaning they have:

(23) Meaning Types of Non-quality Verbs with -iv(i)/-iv(u)

 verb complex


       meaning types




natural process in progress




completed act in progress




effective act in progress




completed process and its result

He interpreted that this aspect form makes continuous process of the head verbs and called it as "vid prodoldzennogo dejstvija" in the sense of continuous aspect in English (p.76). However, it is not correct enough to regard the verb complexes above simply as durativization as a whole. Every example has its special problem. Even the first example with the intransitive verb qHavu-  is not at all easy to interprete. qHavu- is a transitive verb and has the intransitive counterpart qHav-, which is a quality verb with the meaning "to be hot". It has a phonological variant Xavu-dJ which is translated into Russian perfective verb nagret; (make hot) by Panfilov himself. He translated "caj qHavudJ " into "haj nagret". It is not clear whether he showed an imperfective sentence or only one suitable sample of a perfective transitive usage. But he translated the sentence with the aspect marker -ivi (14a) with qHavu-ivi-dJ into "haj nagrevaetsq", namely, in imperfective passive sentence. The question is, whether the original qHavu- is perfective in the meaning "to make something hot". If yes, the sentence (14a) has to be interpreted as "the tea has made hot and is now (yet) hot", namely perfective and resultative; Therefore, the sentence represents the resultative state.
 The verb of the second sentence pHr«-ivi-dJ is translated into imperfective "podxodqt" by Panfilov because of the durative marker. But the original verb pHr«-dJ is clearly perfective according to any  lexicon available to me; it is translated into "prijti" or the analogous words. The sentence (14b) is complex, the verb concerned is embedded as the object sentence of imN m«dJÄu  (they heard). As the Nivkh has no grammatical means of consectio temporum, the tense of the matrix and embedded sentences are the same, namely aorist -dJ. But is is not clear whether the embedded verb tHoZ-Ä«t-ivi-dJ  has to be interpreted as "they have come back and are yet comming back" or simply as "they have been coming back". Preferable is here the former intepretation. Then, here again, the verb complex means perfective and durative, or better to say, progressive activity and its process.
 The example (14c) is the adnominal usage of the verb l«ttJ  (make (up), Russian: delat;, proiz-vodit;). It represents an effective human act, i.e. it belongs to the socalled accomplishment verbs according to Vendler 1967 and Dowty 1979. With the durative marker -ivi/-ivu, the act is described as in progress. Here we have the typical case of a progressive description of a human act.
 The example (14d) includes a verb complex of two kinds of aspect forms tHoZ-Ä«t-ivi, namely, head verb+CON+DUR. The verb tHoZ- appears in some lexicons in the form t¥Z- with the same meaning: "go off, go out". The verb complex can be interpreted as "the fire has gone out and was warm" or "the fire has been going out and was warm". Both interpretations may be possible. But perhaps the former is preferable. Then, the complex has the temporal meaning like perfective and resultative state.  As we have seen above, the aspect form -ivi/-ivu does not always produce the same durative meaning, but it differs according to the LA of the verbs which precede it. This is easy to understand  if we compare (14b) and (14c). Let us think about the example (14c), the sentence refers to an event that at the time t the speaker saw a man making a ship for the time being T. The time T includes the event time t, and the event was reported at the speech time t0. Therefore, the durative meaning comes about from the time relation tT and the tense is established by the time relation tt0. In this sentence, -ivi/-ivu makes up the long interval T out of the verb l«ttJ. This is only possible if the verb has a durative situation inherently. So, this verb must have a LA like #…dur…#, where # indicates a closed frame. On the contrary, the verb complex of (14b) is inherently perfective pHr«dJ . Though the verb is aroist finite -dJ, the sentence does not mean that the hunters have already come back, but they were coming back when the village heard the news. The process of coming back is streched out towards the time when they heard the news by means of -ivi/-ivu which represents a durative or, better to say, a progressive state. But if pHr«dJ alone stands with the finite marker -dJ, the sentence has to be interpreted as they have already come back. Let us mark the coming back process with a streched time interval T, then T precedes the time when the village heard the news at t, and this t precedes the speech time t0, namely Ttt0. Interesting is the relation between pHr«dJ and pHr«ividJ : if the latter stands instead the former in the same sentence, the sentence will be øaø«nJivÄu pHr«dJ Äu imN m«dJэÄu (hunters have come back, they heard). In that case, the time of perfective pHr«dJ Äu will stands after the progressive pHr«ividJ , namely tt0.
 The similar time relation is observed in the sentence (14a). (if) caj qHavudJ (he boiled the tea)  caj qHavuividJ (tea is/was boiling) caj qHavudJ  (tea is boiled) caj qHavdJ (tea is/was hot) . In this case, the aorist -dJ covers both the event and the speech time.
 The time relation of the sentence (14d) is difficult to interpret as above mentioned. Crucial is the ambiguity whether the verb complex means the state of the completed process, namely the resultative state or the completing process. This problem will be settled if we think about the behavior of the aspect form -Ä«t a little more precise in 2.5.
 As far as we have seen above, we assume that Nivh non-quality verbs have at leat the LA types in the table (24). The assumption is, of course, yet tentative, especially because we collected the words only from Panfilov's description on aspect forms -ivi/ivu. But important is to see that the table indicates what types of complex aspect meaning comes about when -ivi/ivu is attached to what types of LA. In other words, we can judge what LA the verb has if it represent a certain complex meaning type when it is concatenated with -ivi/ivu. This implies the aspect form can be utilized like a litmus paper to find out the LA of verbs, just as Japanese -te iru can be used to differenciate durative from resultative LA.

(24) LA types of Non-quality verbs

 head verb of the complex

 meaning of verb complex    

 LA of head verbs 


 qHavu (-ivi-dJ ) (boil)

  is/was already boiling



 pHr«(-ivi-dJ-Äu) (come)

had come/been coming back



  l«t'(-ivu-) (make)

     (was) making



 tHoZ(-Ä«t-ivi-dJ ) (go-out)

   already has gone out


  Suppose that the Otaina's quality verbs make a LA type which represent a state of various subclasses, we symbolize it as *DUR* altogether. We can summarize the LA types of the Nivkh verbs as follows:

(25)  Lexical Aspect Types of Nivkh -- first approach--
   (a)  *DUR*: the open stative, the quality verbs according to Otaina,
   (b)  #DUR*, verbs for natural process 
   (c)  #PUN (+dur)*RES, verbs for resultative punctual activity whose process
   can be durativized
   (d)  #DUR#EFF, verbs for a durative action which leaves a certain effect
    necessarily, i.e. the
socalled accomplishment verbs,
   (e)  #PUN*RES, punctual verbs for activity which leaves a result.
       The examples, see (11), (14) and (24).

 There are surely more LA-types of verbs. We will find them in the next sections in relation to the other spectual forms and their concatenation possiblities.

2. 5. The concatenation of non-quality verbs with the aspect forms -Ä«t and others     

 In order to go farther to find out some more LA types of the Nivkh verbs, we will examine now some crucial examples of the aspect form -Ä«t from Panfilov 1965.

(26) Sentence Examples with -Ä«t (PA§40, pp.69-73)
    a. øaø«ønJivÄ-gu pHr«   -Ä«t   -kэ          t«v     l«t   -tJ.
      hunter-PL    come-CON-soon-after house make-FIN
      The hunters, soon after they have come, made a house.
    b. mэr øafq     it    -tJ, "nJi    tvi    -n«   -dox    qHauk-ra,   nJi  s«k c«ø
      our friend  say-FIN "1PL finish-FUT-DIR  NEG-AF  1PL all you
       Our friend said, "I am not yet defeated. I will have killed you all.
    c. imN   pHr«          -Nan  am-Äu h«   bal         laqv    s«k por-Ä«t-t        
   3PL come-back when tiger-PL this mountain around all lie-CON-PART

       When they came back, tigers were all lying around the mountain.
    d. hoqoNan  h«  øa       kummJ. kutra   qHau               -V«t  -ra.
          then    this animal failed.  failing become-nothing-CON-AF
       Then, this animal fell. Falling (it) has become nothing (died).
    e. utku-oGla     mu   Xoru-Xoru-V«t-tJ.
      woman-child ship rock-rock-CON-FIN
      The child (woman) has rocked the ship.
    f. t«   gins-Vu   morHqa-t     humGa   s«k nJ«N nJivx-gu    kHu-t
   this devil-PL survive-PART exist-IF all  our people-PL  kill-PART
      If these devils remain surviving, they kill all our peoples will have eaten them.
      (will have eaten)

 The first sentence (25a) is straight-forward. The verb pHr«- (come) + V«t  means the perfective conclusion of the action of coming. This aspectual meaning is indicated by the particle -kэ (soon after). The sentence (25b) is a remarkable example; the verb kHudJ/xudJ/iVdJ (kill) has the future suffix -n«, which indicates that the event will be accomplished in the future. This says that the aspect form -V«t is not dependent on tense and represents only the finish of an event. Panfilov interpretes this usage of the aspect form as the marker for coherent action toward the participants of the event. In his sense, the sentence means "I will have killed every one of you altogether." The verb complex of the sentence (25c) por-Ä«t-t  hum-dJ-ra (were lying) is worth thinking about a little more. The constituent verb pordJ  is translated in Russian le'a (le'at; = lie) by Panfilov. Note that this Russian verb is stative, which does not seem to be connected with a conclusive aspect. Panfilov interpretes that in this case, the aspect form -V«t is used to indicate a collected action: "the tigers were lying altogehter". But the verb pordJ  seems to be an action verb inherently. For CT1970 writes that the verb is used to represent the action of human being or animals and cites the usages: muin«r pordJ (to lie from sickness), kHr«r pordJ (to lie for a rest). If so, the verb complex has to be interpreted as that an animate action has been done and the result of the action remained (the result is indicated by humdJ ). Then, pordJ  is an action verb with the resultative state in the meaning like le'a (lying). So, Panfilov's translation is good, but the aspectual feature of the head verb is different from the Russian correspondence. The sentence (25d) is a very interesting example; the aspect form -Ä«t  is here attached to the negative verb qHaudJ which follows the affirmative finite verb kutra (fiel), resulting the meaning "fell and became nothing". Panfilov interpretes this case as intensive usage of the aspect form. The negative verb + the conclusive aspect form qHau-Ä«t- has the meaning that the negative state "non-existence" has come about. Therefore, this complex means that from the process of the precedent verb kudra (fell) occurred and the state "exist nothing" has come about. In short, [-Ä«t  +stative] is resultative/conclusive here, too. Let us go to the sentence (25e). According to Panfilov, -Ä«t  in (25e) indicates the sudden stop of the event process the verb represents. Namely, the swinging of the ship has suddenly stopped. The aspect form -Ä«t  means here also the finish of the verb process which is a durative action represented by the duplicated verb stem. Panfilov interpretes the stop is made "suddenly". Perhaps he intended to show that the aspect form -Ä«t  points the situation directly after the finished action. Anyway, the sentence shows that a long process of action is concluded with neither result nor effect. The last sentence (25f) contains the combination of the aspect form -Ä«t  and the marked tense form -n«-, namely inJ-V«t-n«-dJ-Vu-da. This form represents the conclusive action of eating which occurs in the future. This implies that the aspect form is independent from the tense, as is the case in many other languages.
 As we have seen above, the conclusive aspect form -Ä«t can be attached to various types of verbs and realizes different meanings. We summarize them as follows:

(27) Meaning types of -V«t

forms of V+V«t

   V and represented situation




pHr«-(come): concluded action with result




kHu-(kill): concluded of action with effect



por-V«t-t humdJra

por-(lie): concluded action with result state



kutra qHau-V«t-ra

qHau-(no): coming about of a state




XouXou(-(swing): concluded a long action




inJ-(eat): concluded action without no effect (in the future)


  From this table we can conclude that the aspect form -Ä«t realizes five different meanings according to the LAs of the precedent verbs. We summarize as follows:

(28) LA of the verbs in (27)

    time relation



V1[t] +result(state)

   pHr«-(25a), por-(25c)


  V2[t] + effect




   inJ -(25f)  


  V4[t] + long action






 We have got here some new distinctive types of LA by means of the combination of the aspect form -Ä«t.  Panfilov suggests some more criterions to find other LA. The aspect form -tHu for iterative markingis not possible to be attached to quality verbs. He cites some lexical examples of non-quality verbs like juru-tHu-dJ (read many times, read well) vs. jurudJ (read). But this suffix is said to be unproductive, the other morphological means like reduplication are used for the purpose.
 The conventional/habitual aspect marker -x« has a distinctive function for LA. This form can not be attached to quality verbs, so that it is used to distinguish them from non-quality verbs, as remarked by Otaina 1978, too.
 The complex combinations of different aspect forms represent complicated time relation of events.
The sentence (14d) shows the example in which the conclusive/resultative and the durative aspect forms are combinated to yield a resultative state: tHuÄř tHoZ -Ä«t -ivi -dJ  (the fire has gone out and the state remained). Let us take very complicated combinations: 

(29) Complex aspect forms with different LA types (PA78)
   a. nJ«m«k      pHr«          -ba    pHovo       sidJlu  n« -Ä«t -řadJ
     my mother come=back-after own village soon  do-CON-STATE
     Soon after my mother came back to the village, she was already working.
   b. ci    maøgohař       ja lax  pH     -«řp-Ä«t-ku-řadJ-øa?
      you be-strong-become why cloud self-cover-CON-CAUS-STATE-QUESTION
      If you become strong, why the black cloud will have been covering you?

These examples show that the resultative state of the head verbs is indicated with -Ä«t and is streched out by adJ contaminated form of -hadJ (exist). This complex form can be utiulized to determine the resultativeness (RES) of non-quality verbs.


Abbreviations and Symbols:
TI: Time Indicators
LA: Lexical Aspect (inherent in verbs)
t: There is a time t
e: an event
$…$: ambivalent (=undefined) frame of a LA, $ is * or #
*…$: to the right open frame
#…$: to the right closed frame
: establish, (t) e: a event is established at the time t.
AF: affirmative morpheme
EFF: effective situation (after an accomplished action)
FUT: future morpheme
FIN: finite morpheme
PL: plural morpheme
CAB: coming about (aspect for quality verbs)
CON: conclusion/resultativeness (represented by a morpheme or an aspect
form )

CONJ: conjunction morpheme
CONV: conventional process/action
DUR: a durative process
+DUR: a long durative process
HAB: habitual process/action/… 
INCH: inchoative aspect form
NEG: negation morpheme
PART: a particle for adverbal concatenation
PRV: preservation of quality
RES: resultative situation
Main Literature
Dowty 1979: Dowty, D., Word Meaning and Montague Grammar.
 Synthese Library 7.Reidel Publ.

Gruzdeva 1998: Gruzdeva Ekaterina, Nivkh. Languages of the World.
 Lincom Europa.

Kaneko 1995/2003: 金子亨『言語の時間表現』ひつじ書房。改訂2003
(CD available)
Krejnovich 1958: Krejnovih, E.A. Ob Inkorporirovanii v Nivxskogoij Qzyke
About Incorporation of the Language Nivkh). Voprosy Qzykoznaniq 58-6
Krejnovich 1934: Krejnovih, E.A. Nivxskij (Gilqkskij) Qzyk. v Qzyk i Pis;mennost;
        Narodov Sibira,Hast; III. (Nivkh (Gilyak) Language). Nauhno-issledavatel;naq
       Assosiaciq Instituta Narodov Severa CIK,SSSR
Mattissen 2001: Mattissen, J., Dependent-Head Synthesis in Nivkh and its
 Contribution to a
  Typology of Polysynthesis.Uni Koeln
Otaina 1978: Otaina, G.A., Kahestvennye Glagoly v Nivxskom Qzyke
Otaina,G.A., Quality Verbs in Nivkh Language). Izd. Nauka
Punfilov 1965: Panfilov, V.Z., Grammatika Nivxskogo Qzyka (Panfilov, V.Z.,
 Grammar of 
the Language Nivkh). Akademiq Nauka CCCP
Puxta 1995-2003: Puxta, M.N.,Nivxsko-Russkij Razgovornik i Tematiheskij
( Puxta,M.N.,Nivkh-Russian Conversation and Daily-Life Thesaurus).
 ELPR 2002- A2-017, 
2003, ed. by Lok & Kaneko
Saveleva/Taksami 1970: Savel;va, V.N. i H.M.Taksami, Nivxsko-Russkij Slovar;
Nivkh- Russian Dictionary). Izd. Sovetskaq ?nciklopediq 1970
Vendler 1967, Vendler, Z., Linguistics in Philosophy (Ithaca, N.Y.CornellUniv.)