Instructions for Use
The Online Edition of Grammatical Dictionary of Polish (in Polish: Słownik gramatyczny języka polskiego, henceforth SGJP) follows the 1st and the 2nd CD-Rom editions of the dictionary, however it was supplemented with new functions.
When the user opens the program, a window appears, divided into three basic parts. The upper part is the toolbar, below which there are two panes: the left-hand pane shows a list of headwords and the right-hand pane shows the entry.
SGJP’s main menu can be displayed by hovering the cursor over the SGJP logo located in the top left corner. The menu first allows to choose between two main views of the page: Dictionary Entries and Inflection Patterns. Textual descriptive information is divided into sections.
In the toolbar, in addition to the logo and the title, one can find the following: a search (query) box together with the search button (shaped like a check mark), two sorting buttons (but only in the Dictionary Entries page view), a filter button, a settings button and a help button. Their functions are explained later in these instructions.
This is the main page view of the dictionary, it loads automatically.
Beneath the toolbar there are two panes: the left one with the list of headwords (the headwords pane) and the right one with information on a lexeme chosen from the list (the entry pane).
The Entry Pane
The entry pane consists of the following elements:
- the headword bar,
- the bar with general information on the lexeme,
- the inflection table,
- references (optionally).
The Headword Bar
The headword bar may contain two additional elements: information on pronunciation and a short gloss.
The gloss, in the form of a synonym, typical collocation or other information is intended to facilitate the interpretation of the lexeme. Glosses are introduced for lexemes with a homonymic headword, e.g. for two lexemes wywrzeć: ‘wrząc, wygotować się’ [boil over] and ‘podziałać na coś z pewną siłą’ [e.g. wywrzeć wpływ ‘to influence’, wywrzeć wrażenie ‘to make an impression’], and for lexemes that a contemporary user would find hard to understand, e.g. for the lexeme mesz: ‘rodzaj paszy dla koni’ [a kind of fodder for horses]. As for abbreviations and acronyms, it is either explained what they stand for (sometimes the source language is indicated, e.g. for the lexeme NBA [Eng.] National Basketball Association) or explained through translation (e.g. for BA [British Airlines]: Brytyjskie Linie Lotnicze).
General Information about the Lexeme
General information about lexemes may have different degrees of complexity. It is presented in at least two lines.
In the uppermost line, grammatical information (in Polish) is provided about a lexeme’s membership in a particular lexical class (the repertory of grammatical classes is discussed in “Podstawy teoretyczne”). In the current edition of SGJP, the following lexical classes (more or less corresponding to parts of speech) are distinguished:
- rzeczownik [noun], odsłownik [gerund, verbal noun],
- przymiotnik [adjective], imiesłów bierny [passive participle], imiesłów czynny [active participle],
- liczebnik [numeral],
- czasownik [verb] — właściwy [proper] and niewłaściwy [impersonal, literally — improper] (also: predykatyw [predicative]),
- przysłówek [adverb],
- przyimek [preposition],
- (nieodmienne [indeclinable]:) spójnik [conjunction], partykuła [particle], modyfikator deklaratywności [operator of declarativity], operator adsubstantywny [adsubstantive operator], operator trybu [mood operator], człon frazeologizmu [element of a collocation], człon nazwy [element of a name], przedimek [the first element of a surname of the type van Beethoven] etc.,
- skrót [abbreviation],
- prefiks [prefix].
These labels are attributed to particular units described in the dictionary. However, on certain levels, SGJP uses more general, larger lexical classes (e.g. when applying filters).
Some classes are further subdivided into subclasses to which a given lexeme belongs.
In particular, nouns are sometimes marked as belonging to the semantic class of proper names, more precisely to some of its subclasses, e.g. names of persons (first name imię, surname nazwisko, nickname przydomek etc.), geographical names, names of buildings, organizations, companies, typefaces etc., or parts of those names. Nouns which are not marked should be treated by default as common names.
With regard to verbs (czasownik), they are classified into more exact verb subclasses: proper (właściwy) or improper (niewłaściwy) and as transitive (przechodni) or intransitive (nieprzechodni).
Apart from grammatical information in the top line, qualifiers in the form of abbreviations may be displayed. They serve to delimit the domain in which a given lexeme is used (e.g. muz. ‘muzyczne’ [musical] in the case of the verb solmizować [solmizate]), or its stylistic markedness (e.g. pot. niepopr. ‘potoczne niepoprawne’ [common incorrect usage] in the case of one of the homonymic verbs wartać [be worth] or książk. ‘książkowe’ [bookish] for the numeral gros [gross]). Sometimes they give – after the diamond-shaped symbol ‘◊’ – additional grammatical glosses, especially syntax-related, and stylistic glosses, e.g. for the numeral gros: Też z l.p. ‘Też z liczbą pojedynczą’ [also in singular].
In the case of lexemes taken from Doroszewski’s Dictionary of Polish source information is indicated as: [SJPDor.]. Sometimes other information is given, such as etymology, case government for prepositions, or comments on usage.
In the lower line basic grammatical information about the lexeme is displayed, specifically its inflection pattern, indicated by means of symbols (see below). When – instead of a single symbol of inflectional pattern – there is an alternative of several symbols (marked by slashes), they are preceded by buttons; clicking on either of them will make the corresponding set of forms appear in bold type in the inflection table. For instance, the lexeme słabeusz [weakling] – of m1 gender – is provided with two inflectional patterns: B2+(y)/B2+e(ów). The first paradigm includes a genitive plural with the more frequent ending +y, the second one a genitive plural with the ending +ów. The order of presentation is not to be taken as a strict recommendation regarding the usage. If a given inflection pattern for a given lexeme is clearly less frequent or considered incorrect, it is marked with an appropriate comment (e.g. niezal. ‘niezalecane’ [not recommended] for the inflection żebram, żebrają of the verb żebrać [beg]).
Clicking the name of an inflection pattern will open information about the pattern – the page view then changes from Dictionary Entries to Inflection Patterns.
For two classes of lexemes, names of inflection patterns are preceded with important introductory information:
- for nouns – about their gender (this is actually syntactical information, however of a kind that significantly influences the inflection type),
- for verbs – about aspect (which significantly influences the repertory of forms displayed in the conjugation tables).
The main part of the entry pane is the inflection table for the lexeme. It can be presented in two ways, namely as a display of all forms or of base forms. The user can choose one or the other by selecting the preferred version in the Inflection settings, which can be found by hovering the mouse over the Settings button on the toolbar. In the first case we obtain the complete paradigm, i.e. all forms of a particular lexeme; in the second case only the base forms from which all forms of the given lexeme can be made (the choice of base forms for particular grammatical classes is discussed in Podstawy teoretyczne).
In the case of nouns with two or more genders (e.g. chamisko [boor]), program displays the corresponding number of tables, side by side. In the case of uninflected lexemes, the inflection table is limited to one form.
For nouns of m1 gender (masculine personal), cells of nominative (and vocative) plural forms in the inflection table are divided into two subcells, which contain characteristics: ndepr ‘niedeprecjatywna’ [non-depreciative] and depr ‘deprecjatywna’ [depreciative]. For some nouns they present respectively non-depreciative and depreciative forms of that lexeme, e.g. for the lexeme syn [son] – ndepr. synowie and depr. syny (compare: Moi synowie spali and Moje syny spały). The first of those forms – neutral, unmarked – is displayed normally, like forms of other cases, and the other – marked, used with some stylistic intention – is displayed in a less bolded font. The choice of the neutral form of a lexeme depends on how the two forms are used: for the lexeme syn synowie is the neutral form, and for the lexeme sukinsyn [son of a bitch] the neutral form is sukinsyny (this lexeme has two non-depreciative variants: sukinsyni and sukinsynowie).
The last part of the entry pane contains references with information about word-formation: the most regular derivatives of the lexeme (or the opposite: information about the basis from which the lexeme was derived). The most important types of derivatives introduced in SGJP are: for verbs – verbs of opposing aspect in an aspectual pair (e.g. robić [do] has a reference link to zrobić), adjectival participles (e.g. active robiący [doing] and passive robiony [done]), as well as verbal nouns (e.g. robienie [doing]); for adjectives – adverbs and nouns ending in -ość denoting features (e.g. słaby [weak] has reference links to słabo [weakly] and słabość [weakness]), as well as comparatives (e.g. słabszy [weaker]). With regard to surnames, as well as some other subgroups of people’s names, among the references there are links to masculine or feminine forms (e.g. Kowalski – Kowalska). For geographical and topographical names, one can find there links to adjectives derived from the lexeme (e.g. Warszawa – warszawski). References given in entry panes are not rigorous; they only serve immediate practical purposes. So, aspectual counterparts are presented only in simple cases and usually for one of the meanings of the given verb (aspectual pairs are presented primarily with a view to foreigners using SGJP).
The references are actually hyperlinks. By clicking on an item visible as a reference, one obtains dictionary information about the lexeme it represents.
The Headwords PaneThe pane containing the list of headwords is provided with a vertical scrollbar, which allows the user to scroll the list.
Headwords on the list are displayed along with a stripe that varies in colour, located to the right of them. The colour of the stripe indicates the lexeme's class. Headwords of verbs (e.g. abdykować [abdicate]), regular derivatives of verbs like verbal nouns (e.g.abdykowanie) and adjectival participles (e.g. abdykujący) are marked with an orange stripe. A green stripe marks an item as an adjective (e.g. zielony [green]), as a regular deadjectival derivative like an adverb (e.g. zielono) or a deadjectival characteristic (e.g. zieloność). The blue stripe marks an item as a noun, purple marks it as a numeral and grey marks it as belonging to the residual, or "other", class.
The colours in their function of associating headwords with lexeme’s classes are also present in the entry pane – they are used as the colour of lines of the inflection table.
To the right of the coloured stripe there is additional information about the lexeme. The user can decide which elements to display there. He can choose one or more of four attributes:
- lex. cl. ‘lexical class'
- gnr. (gender for nouns, aspect for verbs)
- qual. ‘qualifier'
The content of the headwords list can be adapted to the needs of users via Filters. Filter button (in the shape of the funnel) is located in right-hand side of the toolbar.
In the filtering window that shows up after clicking the funnel symbol, the user can choose in the top-left box one of two ways of combining filtering criteria: disjunctive (or) or conjunctive (and). Right next to it, there is a button with “+” which allows the user to add other criteria; all criteria added by the user are joined with the same conjunction.
Filtering criteria that show up in the successive lines (after clicking the button “+”) can be divided into open and closed ones. When using closed criteria, the user chooses from the possibilities given in the list (the list of values is presented; they have to be specified by either equal to and not equal to criteria), while using open criteria, it is necessary to type in the sequence of characters that should be filtered.
Open criteria are more important because they filter sequences of characters, especially letters. The user has to type a string of characters that will be filtered. Typed-in string can serve as the base for searching in SGJP for units, which are equal to or not equal to the typed sequence, which start with or don’t start with the typed sequence, which end with or don’t end with the typed sequence, or which contain or don’t contain the typed sequence in them. Moreover, the searched words may match the pattern or not. The entered pattern is a so-called regular expression.
Regular expressions are used in computer science applications. They allow to formally write a string of characters (expression) in a generalized way. Such notation follows various systems, elements of which in considerable number of cases are similar. Presenting the convention applied in SGJP would overload this instruction, so only the most basic examples are presented. Assume we search for Polish words consisting of three letters, second of which being “ó” (e.g. móc [be able to], dół [pit], pół [half]). The user will get the demanded list of words after typing the following pattern: ^.ó.$ where the . sign corresponds to any letter, ^ – the beginning of the string of characters, and $ – the end of the string. The "(not) matched by pattern" filtering criterion is formulated somewhat confusingly, since the regular expression is matched to any substring of the typed-in unit, not to the whole; in fact, it should be: "(doesn’t) contain a string matching the pattern". In the result of matching to the chosen criterion, the list of 36 strings of characters is displayed (bób [broad bean], bóg [god] etc.). If the typed pattern was modified to: ^.*ó.$ (the asterisk * sign, placed after a character, corresponds with any number of characters), the list will expand to 1063 results and will include such words as: król [king], Kraków [Cracow] etc. Using regular expressions is recommended to more advanced users, who are skilled in operating on strings of characters.
In the similar way it is possible to filter by Form criterion. Applying this filter will search for words which correspond to any form occurring in any inflection paradigm of any lexeme in SGJP. While searching for words that match by pattern ^.ó.$, the list of 88 strings of characters (headwords) will be displayed. Aside from all words obtained previously (that are the result of applying Headword filter), there are there such lexemes as: robić [do] (because of its imperative form: rób) or poła [coat-tail] and póła [big shelf] (because of their genitive plural form: pół). If we apply a modified pattern: ^.*ó.$, the list will grow to 40024 results (because of forms like tyłów [back] Gen.Pl. of tył)
Other open criteria that theoretically give the same possibility of matching the typed-in sequences, Pattern and Pronunciation, are considerably less useful. They allow to search through the sets of names of patterns and transcripts of pronunciation. Last two open criteria, Pattern count and Gender count, allow to specify the count with either equal to, greater than or less than criteria.
Closed criteria are as follows: Lexical class, Pattern type, Commonness, Qualifier, Gender, Reflexivity, Aspect. The repertory of values of those parameters should be clear to the user after displaying the list of possible options. Only first three closed criteria require explaining.
In the case of the Lexical class criterion, the user has to choose from the following values (abbreviated symbols in the brackets correspond with the values of lex.cl. attribute in the list of headwords):
- czasownik [verb] (cz.)
- imiesłów bierny [passive participle] (im. b.)
- imiesłów czynny [active participle] (im. cz.)
- liczebnik [numeral] (licz.)
- nieodmienne [indeclinable] (ndm)
- odsłownik [gerund, verbal noun] (ods.)
- predykatyw [predicative] (pred.)
- prefiks [prefix] (pref.)
- przyimek [preposition] (pim.)
- przymiotnik [adjective] (przym.)
- przysłówek [adverb] (ps.)
- rzeczownik [noun] (rz.)
- skrót [abbreviation] (skr.)
The Pattern type criterion allows to choose a value of the parameter which is not present in the lexeme’s description at all. Names and types of patterns were discussed in the separate article Systematyzacja wzorów, available in Polish in SGJP.
The use of Commonness criterion is quite complex. To filter with this criterion, you have to choose from the following values (which are of equal rank):
- nazwa pospolita [common noun]
- nazwa własna [proper noun]
- człon nazwy własnej [a part of a proper noun]
- marka [brand name]
- nazwa geograficzna [geographical name'
- człon nazwy geograficznej [a part of a geographical name]
Combining Filtering Criteria
Filters can be combined disjunctively or conjunctively by clicking the button with “+” sign. However, with regard to closed criteria, it is possible to combine filtering in yet another way (to some extent): since they require to choose (tick) values from the list, it is possible to choose several different values at the time.
For example, to get the list of all words in SGJP qualified as proper names, the user has to – beside the proper name value – choose all subclasses of that criterion.
Similarly, by choosing several values in Lexical class criterion, it is easy to display on the list of headwords only large classes of lexemes which have similar grammatical properties, e.g. lexemes of substantive inflection type (nouns and verbal nouns) or adjectival inflection type (adjectives, present participles, and past participles). The list of verbs (in the broad sense) will be shown after choosing from the list both verbs and predicatives.
The Order of Headwords
The headwords list can be arranged in two ways: the ordinary alphabetical order (lat. a fronte), and the reverse alphabetical order (lat. a tergo). The current order can be changed by pressing one of the sorting buttons to the right of the query field in the toolbar.
When sorted in a tergo order (i.e. backwards, in the reverse alphabetical order, where the last letter has the highest rank and the preceding letters have successively decreasing ranks), lexemes with similar inflection are in close proximity. This sorting order may also be helpful while matching words that rhyme.
The list of headwords can be browsed with the vertical scrollbar. There are several ways to find an entry. The easiest way is to click the headword in the list. It is also possible to use Filters, applying Headword – equals to – ... criterion. Headwords can also be search by typing them into the query field: if the user will not start the search by pressing either [Enter] or the search button, located directly next to the query field (in the shape of the tick), the headword list will automatically move to the entered headword, and the entry pane will display information on searched entry.
This mechanism works for not only headwords, but also words (strings of characters): after typing a word, SGJP will search for headwords of the closest resemblance to the typed-in word, following the selected sorting order (a fronte or a tergo).
For example, in a fronte order of the headwords in the list, in the course of typing the successive letters of a headword, the first lexeme to be displayed in the entry pane will be an entry whose initial part contains this particular sequence of letters. E.g, while typing zabawkarstwo [toy making], the entry pane will display consecutively: z, za, zababrać, zabawa, zabawka, zabawkarski, zabawkarstwo.
If the user types in the query field the word that does not correspond to the existing form of lexeme present in SGJP, it will be matched to the strings of characters of the closest resemblance listed in the headwords pane. For example, if the user enters the string etla and confirms their search, in a fronte order the dictionary will find the lexeme Etna, and in a tergo order basetla [double-bass], because the dictionary matches typed sequence to the ending of the headword form.
The headwords are always single words. Their order will not be affected by any non-letter character occurring within them. For example: the entry dum-dum in the ordinary list is placed between dumca and dumiec, and in the reverse list – between oppidum a memorandum.
It is also possible to search for a lexeme on the basis of any of its forms. SGJP does not search a given string through the list of headwords, but through the full list of words belonging to all lexemes, when the user accepts the query by pressing the key [Enter] or clicking the search button. For example, introducing the word ód in one of two aforementioned ways will result in showing the headword oda [ode] in the headwords pane and display information about this lexeme in the entry pane (ód is its Gen.Pl.).
SGJP has also an additional function: if the typed word is homonymous, i.e. can be interpreted as a form of several different lexemes, only one of its interpretations is presented in the entry pane; however, after pressing the key Enter or the search button, an additional small window will be displayed in the upper part of the headwords list. This window will comprise “a sublist of proposed lexemes”, all of which contain the typed word as one of their forms (e.g. for the word żółci the sublist will encompass the lexemes: żółcić, żółć, żółty, and for the word woli — lexemes: wola, wole, woleć, woli). After choosing one of the lexemes, a corresponding entry will be displayed.
Inflection patterns page view was designed for the purposes of Online Edition of SGJP. It was based on the Dictionary entries page view.
Beneath the toolbar there are two panes: the left pane which displays the list of patterns (patterns pane) and the pane which displays information on chosen pattern (information pane).
The Information Pane
In the right pane an information on given pattern is displayed, namely:
- the name of the pattern
- the supertype and type of the pattern
- information on lexemes belonging to the given pattern
- table of base forms
The name of the pattern is a fundamental identifying element. The system of patterns was described in a separate article: Systematyzacja wzorów.
Patterns presented in SGJP have various degrees of complexity. In the result, they are divided into types and supertypes (in the dictionary: infl. types [inflection types]), which determine the general morphologic profile: substantival – subst., pronominal – ppron., adjectival – adj., numeral – num., verbal – v., predicative – pred., indeclinable – ndm. ‘nieodmienny’, and abbreviational – skr.. Information on every type includes the scheme of base forms. Base forms have different labels assigned, indicating their place in paradigms.
Total lexeme count makes a total of lexemes that belong to given type of pattern; Lexeme count, located below – counts the amount of lexemes that belong to important subclasses of the given class (nouns of a given gender, verbs of a given aspect) with examples for each of them. Both numbers and examples are highlighted as hyperlinks. Clicking an example will result in opening the Dictionary entries page view of the given lexeme; clicking the numbers – in opening the Dictionary entries page view of the example and displaying in the left pane the list of lexemes, which inflect following the chosen pattern.
The table with base forms assigns labels to the corresponding forms of a chosen example. The repertory of base forms for particular types was discussed in the theoretical introduction. Words, corresponding to the forms in the table with given labels, were divided into two parts with the ∙ sign; which separates the stem, i.e. common part for all forms of example lexeme, and the ending, i.e. the part that differentiate base forms from each other. Other lexemes of the same inflection pattern have other stems, but the same set of endings.
The Patterns PaneThe pane with the list of patterns works similarly to the list of headwords, but it contains names of patterns instead (it’s recommended to become familiar with the labelling system before using the view of patterns – see: symbols and notation). Likewise, there are also stripes of different colours to the right of pattern names (the meaning of the colours are the same throughout the whole SGJP dictionary). Further to the right, one can also find additional information on a pattern. The user can choose between three elements to be displayed in the list by selecting them in the subsidiary table. The table can be displayed after clicking Pattern list in the settings. The user can select the following available pattern’s attributes:
- infl. type ‘inflection type’ (or supertype),
The content of the list of patterns, like the list of headwords, can be adjusted to the user’s needs by filters. It is also possible to combine filtering criteria (disjunctively or conjunctively) and to create alternative values in case of closed criteria.
The patterns can be filtered with three criteria: the name, the inflection type (supertype) and the type. Those parameters were explained earlier in the instruction. The first criterion is open (the user has to enter the pattern’s name) and both second and third – closed (the user has to choose one of two given options).
Settings button allows to choose the following elements of the view:
- (only for the lexeme page view) the way of displaying the paradigm: all forms or base forms
- information to be displayed in the list of headwords (lex. cl., gnr., patterns, qual.) or in the list of patterns (type, infl. type, example)
- site’s language (English, Polish)
Clicking the help button displays descriptive information, which is also available for the user in the main menu of SGJP.