Definitions

done deed

Hungarian language

Hungarian (magyar nyelv ) is a Uralic language (more specifically a Ugric language) unrelated to most other languages in Europe. It is spoken in Hungary and by the Hungarian minorities in seven neighbouring countries. The Hungarian name for the language is magyar (ˈmɒɟɒr̪).

Hungarian has long been of great interest to linguists as one of the small number of modern European languages that do not belong to the Indo-European language family. Due to the Uralic heritage, Hungarian often sounds completely foreign to speakers of Indo-European languages. It is commonly considered to be one of the most difficult languages for speakers of English (or other Indo-European languages) to learn well.

There are about 14.5 million native speakers, of whom 9.5-10 million live in modern-day Hungary. A further two million speakers live outside Hungary in areas that were part of the Kingdom of Hungary before the Treaty of Trianon. Of these, the largest group lives in Romania, where there are approximately 1.4 million Hungarians (see Hungarian minority in Romania). Hungarian-speaking people are also to be found in Slovakia, Serbia, Ukraine, Croatia, Austria, and Slovenia, as well as about a million people scattered in other parts of the world (see Geographic distribution). As with many European languages, there are a few hundred thousand speakers of Hungarian in the United States as well.

History

Classification

Hungarian is a Uralic language, more specifically a Ugric language and the most closely related languages are Mansi and Khanty of western Siberia. Connections between the Ugric and Finnic languages were noticed in the 1670s and established, along with the entire Uralic family, in 1717, although the classification of Hungarian continued to be a matter of political controversy into the 18th and even 19th centuries. Today the Uralic family is considered one of the best demonstrated large language families, along with Indo-European and Austronesian. The name of Hungary could be a corruption of Ungrian/Ugrian, and the fact that the Eastern Slavs referred to them as Ǫgry/Ǫgrove (sg. Ǫgrinŭ) seemed to confirm that. As to the source of this ethnonym in the Slavic languages, current literature favors the hypothesis that the Turkic "On-ogur" ("Ten arrows" or "Ten tribes") is the origin for the word Hungarian .

There are numerous regular sound correspondences between Hungarian and the other Ugric languages. For example, Hungarian /aː/ corresponds to Khanty /o/ in certain positions, and Hungarian /h/ corresponds to Khanty /x/, while Hungarian final /z/ corresponds to Khanty final /t/. For example, Hungarian ház (haːz) "house" vs. Khanty xot (xot) "house", and Hungarian száz (saːz) "hundred" vs. Khanty sot (sot) "hundred".

The distance between the Ugric and Finnic languages is greater, but the correspondences are also regular.

See also: Regular sound correspondences between Hungarian and other Uralic languages

Antiquity and the early Middle Ages

As Uralic linguists claim, Hungarian separated from its closest relatives approximately 3000 years ago, so the history of the language begins around 1000 BC. The Hungarians gradually changed their way of living from settled hunters to nomadic cattle-raising, probably as a result of early contacts with Iranian nomads. Their most important animals included sheep and cattle. There are no written resources on the era, thus only a little is known about it. However, research has revealed some extremely early loanwords, such as szó ('word'; from the Turkic languages) and daru ('crane', from the related Permic languages.)

The Turkic languages later, especially between the 5th and the 9th centuries, had a great influence on the language. Most words related to agriculture, to state administration or even to family relations have such backgrounds. Interestingly, Hungarian syntax and grammar was not influenced in a similarly dramatic way.

The Hungarians migrated to the Carpathian Basin around 896 and came into contact with Slavic peoples – as well as with speakers of Romance languages –, borrowing many words from them (for example tégla – "brick", mák – "poppy", or karácsony – "Christmas"). In exchange, the neighbouring Slavic languages also contain some words of Hungarian origin (such as Croatian čizma (csizma) – "boot", or Serbian ašov (ásó) – "spade").

The first written accounts of Hungarian, mostly personal and place names, are dated back to the 10th century. Hungarians also had their own writing system, the Old Hungarian script, but no significant texts remained from the time.

Since the foundation of the Kingdom of Hungary

The Kingdom of Hungary was founded in 1000, by Stephen I of Hungary (Hungarian:I.(Szent) István király). The country was a western-styled Christian (Roman Catholic) state, and Latin held an important position, as was usual in the Middle Ages. Additionally, the Latin alphabet was adopted to write the Hungarian language.

Therefore, Hungarian was also heavily influenced by Latin. The first extant text of the language is the Funeral Sermon and Prayer, written once in the 1190s. The earliest example of Hungarian religious poetry is the Old Hungarian 'Lamentations of Mary', a poem about the afflictions of Mary when she saw the death of her son. More extensive literature in the Hungarian language arose after 1300. The first Bible translation is the Hussite Bible from the 1430s.

The language lost its diphthongs, and several postpositions transformed into suffixes, such as reá 'onto' 1055: utu rea 'onto the way'; later: útra). Vowel harmony was also developed. At one time, Hungarian used six verb tenses; today, only two (the future not being counted as one, as it's a compound formed with an auxiliary verb).

The first printed Hungarian book was published in Cracow in 1533, by Benedek Komjáti. The work's title is Az Szent Pál levelei magyar nyelven, i.e. The letters of Saint Paul in the Hungarian language. In the 17th century, the language was already very similar to its present-day form, although two of the past tenses were still used. German, Italian and French loans also appeared in the language by these years. Further Turkish words were borrowed during the Ottoman occupation of much of Hungary between 1541 and 1699.

In the 18th century, the language was incapable of clearly expressing scientific concepts, and several writers found the vocabulary a bit scant for literary purposes. Thus, a group of writers, most notably Ferenc Kazinczy, began to compensate for these imperfections. Some words were shortened (győzedelem > győzelem 'triumph'); a number of dialectal words spread nationally (e. g. cselleng 'dawdle'); extinct words were reintroduced (dísz 'décor'); a wide range of expressions were coined using the various derivative suffixes; and some other, less frequently used methods of expanding the language were utilized. This movement was called the 'language reform' (Hungarian: nyelvújítás), and produced more than ten thousand words, many of which are used actively today. The reforms led to the installment of Hungarian as the official language over Latin in the multiethnic country in 1844.

The 19th and 20th centuries saw further standardization of the language, and differences between the mutually already comprehensible dialects gradually lessened. In 1920, by signing the Treaty of Trianon, Hungary lost 71% of its territories, and along with these, 33% of the ethnic Hungarian population. Today, the language is official in Hungary, and regionally also in Romania, Slovakia, and Serbia.

Geographic distribution

Hungarian is spoken in the following countries as a mother tongue:

Country Speakers
Hungary 10 million (census 2001)
Romania
(mainly Transylvania)
1,443,970 (census 2002)
Slovakia 520,528 (census 2001)
Serbia
(mainly Vojvodina)
293,299 (census 2002)
Ukraine
(mainly Zakarpattia)
149,400 (census 2001)
United States 117,973 (census 2000)
Canada 75,555 (census 2001)
Israel 70,000
Austria
(mainly Burgenland)
22,000
Croatia 16,500
Slovenia 9,240
Total 12,5-13 million (in Pannonian Basin)
Source: National censuses, Ethnologue

About a million more Hungarian speakers live in Argentina, Australia, Belgium, Brazil, the Czech Republic, Finland, France, Germany, The Netherlands, Italy, Switzerland, the United Kingdom, the United States, Venezuela, and in other parts of the world.

Official status

Hungarian is the official language of Hungary, and thus an official language of the European Union. Hungarian is also one of the official languages of Vojvodina and an official language of three municipalities in Slovenia: Hodoš, Dobrovnik and Lendava, along with Slovene. Hungarian is officially recognized as a minority or regional language in Austria, Croatia, Romania, Bukovina, Zakarpattia in Ukraine, and Slovakia. In Romania and Slovakia, it is an official language at local level in all communes, towns and municipalities with an ethnic Hungarian population of over 20%.

Dialects

The dialects of Hungarian identified by Ethnologue are: Alföld, West Danube, Danube-Tisza, King's Pass Hungarian, Northeast Hungarian, Northwest Hungarian, Székely and West Hungarian. These dialects are, for the most part, mutually intelligible. The Hungarian Csángó dialect, which is not listed by Ethnologue, is spoken mostly in Bacău County, Romania. The Csángó minority group has been largely isolated from other Hungarians, and they therefore preserved a dialect closely resembling medieval Hungarian.

Phonology

Hungarian has 14 vowel phonemes and 25 consonant phonemes. The vowel phonemes can be grouped as pairs of long and short vowels, e.g. o and ó. Most of these pairs have a similar pronunciation, only varying in their duration; the pairs /<á> and /<é> differ both in closedness and length, however.

Consonant phonemes of Hungarian
  Bilabial Labio-
dental
Alveolar Post-
alveolar
Palatal Velar Glottal
Nasal m n ɲ
Plosive p  b t  d k  ɡ
Affricate t͡s  d͡z t͡ʃ  d͡ʒ c͡ç  ɟ͡ʝ
Fricative f  v s  z ʃ  ʒ h   
Trill r
Approximant l j

Consonant length is also distinctive in Hungarian. Most of the consonant phonemes can occur as geminates.

The sound voiced palatal plosive /ɟ/, written , sounds similar to 'd' in British English 'duty' (in fact, more similar to 'd' in French 'dieu'). It occurs in the name of the country, "Magyarország" (Hungary), pronounced /ˈmɒɟɒrorsaːg/.

Primary stress is always on the first syllable of a word, as with its cousin Finnish and neighboring languages, Slovak (Standard dialect) and Czech. There is sometimes secondary stress on other syllables, especially in compounds, e.g. viszontlátásra ("goodbye") pronounced /ˈvisontˌlaːtaːʃrɒ/.

Front-back vowel harmony is an important feature of Hungarian phonology. See the Hungarian phonology article for more details.

Single /r/s are tapped, like the Spanish pero; double /r/s are trilled, like the Spanish perro.

Grammar and syntax

Hungarian is an agglutinative language – it uses a number of different affixes, including suffixes, prefixes and a circumfix to define the meaning or the grammatical function. Instead of prepositions, which are common in English, Hungarian uses only postpositions.

There are two types of article in Hungarian:

  • definite: a before words beginning with consonants and az before vowels (in a phonological sense, behaving just like the indefinite article ’a(n)’ in English)
  • indefinite: egy, literally ‘one’.

Nouns have as many as eighteen cases. Of these, some are grammatical, e.g. the unmarked nominative (for example, az alma ‘the apple’), and the accusative marked with the suffix –t (az almát). The latter is used when the noun in question is used as the object of a verb. Hungarian does not have a genitive case, and numerous English prepositions are equivalent not to an affix, but to a postposition, as in az alma mellett ‘next to the apple’. Plurals are formed using the suffix –k (az almák ‘the apples’). Adjectives precede nouns, e. g. a piros alma ‘the red apple’. They have three degrees, including base (piros ‘red’), comparative (pirosabb ‘redder’), and superlative (legpirosabb ‘reddest’). If the noun takes the plural or a case, the adjective, used attributively, does not agree with it: a piros almák ‘the red apples’. However, when the adjective is used in a predicative sense, it must agree with the noun: az almák pirosak ‘the apples are red’. Adjectives also take cases when they are used without nouns: Melyik almát kéred? - A pirosat. 'Which apple would you like? - The red one.'

Verbs developed a complex conjugation system during the centuries. Every Hungarian verb has two conjugations (definite and indefinite), two tenses (past and present-future), and three moods (indicative, conditional and imperative), two numbers (singular or plural), and three persons (first, second and third). Out of these features, the two different conjugations are the most characteristic: the "definite" conjugation is used for a transitive verb with a definite object. The "indefinite" conjugation is used for an intransitive verb or for a transitive verb with an indefinite object. These rules, however, do not apply everywhere. The following examples demonstrate this system:

John lát. ‘John can see.’
(indefinite: he has the ability of vision)
John lát egy almát. ‘John sees an apple.’
(indefinite: the apple can be any of the world’s apples)
John látja az almát. ‘John sees the apple.’
(definite: John sees a specific apple)
See also: Definite and indefinite conjugations.

Present tense is unmarked, while past is formed using the suffix –t or sometimes –tt: lát 'sees'; látott 'saw', past. Futurity is often expressed with the present tense, or using the auxiliary verb fog ‘will’. The first most commonly applies when the sentence also defines the time of the future event, for example John pénteken moziba megy – literally ‘John on Friday into cinema goes’, i.e. ‘On Friday, John will go to the cinema.’ In the other case, the verb’s infinitive (formed using –ni) and the ‘fog’ auxiliary verb is used: John moziba fog menni – ‘John will go to the cinema.’ This is sometimes counted as a tense, especially by non-specialist publications.

Indicative mood is used in all tenses; the conditional only in the present and the past, finally the imperative just in the present. Indicative is always unmarked. Verbs also have verbal prefixes. Most of them define movement direction (lemegy – goes down, felmegy – goes up), but some of them give an aspect to the verb, such as the prefix meg-, which defines a finite action.

Hungarian word order is often mentioned as free, i.e. because of marking the object using –t, it is not always necessary to place the subject before the verb, and the object after it, as in English. This feature makes Hungarian able to focus on particular sections of the sentence – generally, the beginning of the sentence contains the most important information:

John lát egy almát. ‘John sees an apple.’
(when it is important to stress that it's John, not someone else, who sees an apple; or when no special stress is required)
John egy almát lát. (or even Egy almát lát John) ‘John sees an apple.’
(when it is important that it's an apple John sees, and not something else. The same emphasis could be translated as 'What John sees is an apple.')

Politeness

There exists a four-tiered system for expressing levels of politeness in the Hungarian language. It is tangentially similar to the Japanese notion in purpose.

  • Ön (önözés): Use of this form in speech shows respect towards the person addressed, but it is also the common way of speaking in official texts and business communications. Here "you", the second person, is grammatically addressed in the third person.
  • Maga (magázás, magázódás): Use of this form serves to show that the speaker wishes to distance himself/herself from the person he/she addresses. A boss could also address a subordinate as "maga". Aside from the different pronoun it is grammatically the same as "önözés".
  • Néni/bácsi (tetszikezés): Children are supposed to address adults they're not close friends with using "tetszik" ("you like") as some kind of an auxiliary verb with all other verbs. "Hogy vagy?" ("How are you?") here becomes "Hogy tetszik lenni?" ("How do you like to be?"). The elderly are generally addressed this way, even by adults. Whoever uses this way of speaking will not use normal greetings, but can only say "(kezét) csókolom" ("I kiss (your hand)"). This way of speaking is perceived as somewhat awkward and often creates impossible grammatical structures, but is still widely in use.
  • Te (tegezés, tegeződés or pertu, per tu from latin): Used generally, i.e. with persons with whom none of the above forms of politeness is required. Interestingly, the highest rank, the king was traditionally addressed "per tu" by all, be it a peasant or a nobleman, although Hungary not having any crowned king since 1918 this notion survives only in folk tales and children's stories. Use of "tegezés" in media and advertisements has become more frequent since the early 1990s. The growing significance of English language media, especially the internet also supports the expansion of using "tegezés".

The four-tiered system was already somewhat eroded between 1947 and 1989, when communist rule mandated that all people call each other "elvtárs" (comrade) and the current expansion of "tegeződés" may also threaten its long-term survival.

Lexicon

Example with ad
Hungarian English
Derived terms
ad gives
adás transmission
adó tax
adóhivatal tax/revenue office
adózik pays tax
adózó taxpayer
adós debtor
adósság debt
adalék additive (ingredient)
adag dose, portion
With verbal prefixes
megad repays (debt)
eladó for sale, salesperson
hozzáad augments, adds to
As part of compounds
rádióadó radio station/radio transmitter
adomány / from the Latin
dominum=dominyum
word integration/
donation
adoma anecdote
Giving an exact estimate for the total word count is difficult, since it is hard to define what to call "a word" in agglutinating languages, due to the existence of compound words. To have a meaningful definition of compound words, we have to exclude such compounds whose meaning is the mere sum of its elements. The largest dictionaries from Hungarian to another language contain 120,000 words and phrases (but this may include redundant phrases as well, because of translation issues). The new desk lexicon of Hungarian language contains 75,000 words and the Comprehensive Dictionary of Hungarian Language (to be published in 18 volumes in the next twenty years) will contain 110,000 words. The default Hungarian lexicon is usually estimated to comprise 60,000 to 100,000 words. (Independently of specific languages, speakers actively use at most 10,000 to 20,000 words, with an average intellectual using 25-30 thousand words.) However, all the Hungarian lexemes collected from technical texts, dialects etc. would all together add up to 1,000,000 words.

Hungarian words are built around so-called word-bushes. (See an example on the right.) Thus, words with similar meaning often arise from the same root.

The basic vocabulary shares a couple of hundred word roots with other Uralic languages like Finnish, Estonian, Mansi and Khanty. Examples of such include the verb él 'live' (Finnish elä), the numbers kettő 'two', három 'three', négy 'four' (cf. Mansi китыг kitig, хурум khurum, нила nila, Finnish kaksi, kolme, neljä, Estonian kaks, kolm, neli, ), as well as víz 'water', kéz 'hand, arm', vér 'blood', fej 'head' (cf. Finnish and Estonian vesi, käsi, veri, Finnish pää, Estonian pea or 'pää'').

The proportion of the word roots in Hungarian lexicon is as follows: Finno-Ugric 21 %, Slavic 20 %, German 11 %, Turkic 9.5 %, Latin and Greek 6 %, Romance (mostly Romanian) 2.5 %, other of known origin 1 %, other of uncertain origin 30%. Except for a few Latin and Greek loan-words, these differences are unnoticed even by native speakers; the words have been entirely adopted into the Hungarian lexicon. There are an increasing number of English loan-words, especially in technical fields.

Word formation

Words can be compound (as in German) and derived (with suffixes).

Compounds

Compounds are present since the Proto-Uralic era in the language. Numerous ancient compounds transformed to base words during the centuries. Today, compounds play an important role in vocabulary.

One good example for these is the word arc:

orr (nose) + száj (mouth) → orca (face) (colloquial until the end of the 19th century and still in use in some dialects) → arc (face)

Compounds are made up of two base words: the first is the prefix, the latter is the suffix. A compound can be subordinative: the prefix is in logical connection with the suffix. If the prefix is the subject of the suffix, the compound is generally classified as a subjective one. There are objective, determinative, and adjunctive compounds as well. Some examples are given below:

Subjective:
menny (heaven) + dörög (thunder) → mennydörög (thundering)
nap (Sun) + sütötte (baked) → napsütötte (sunlit)
Objective:
fa (tree, wood) + vágó (cutter) → favágó (lumberjack, literally "woodcutter")
Determinative:
új (new) + (modification of -vá, -vé a suffix meaning "making it to something") + építés (construction) → újjáépítés (reconstruction, literally "making something to be new by construction")
Adjunctive:
sárga (yellow) + réz (copper) → sárgaréz (brass)

According to current orthographic rules, a subordinative compound word has to be written as a single word, without spaces; however, if the length of a compound of three or more words is over six syllables, a hyphen must be inserted at the appropriate boundary to avoid ambiguity.

Other compound words are coordinatives: there is no concrete relation between the prefix and the suffix. Subcategories include word duplications (to emphasise the meaning; olykor-olykor 'really occasionally'), twin words (where a base word and a distorted form of it makes up a compound: gizgaz, where the suffix 'gaz' means 'weed' and the prefix giz is the distorted form; the compound itself means 'inconsiderable weed'), and such compounds which have meanings, but neither their prefixes, nor their suffixes make sense (for example, hercehurca 'long-lasting, frusteredly done deed').

A compound also can be made up by multiple (i.e., more than two) base words: in this case, at least one word element, or even both the prefix and the suffix is a compound. Some examples:

elme [mind; standalone base] + (gyógy [medical] + intézet [institute]) → elmegyógyintézet (asylum)
(hadi [militarian] + fogoly [prisoner]) + (munka [work] + tábor [camp]) → hadifogoly-munkatábor (work camp of prisoners of war)

Noteworthy lexical items

Points of the compass

Hungarian words for the points of the compass are directly derived from times of day.

  • North = észak (from "éj(szaka)", 'night'), as the Sun never shines from the North
  • South = dél ('noon'), as the Sun shines from the South at noon
  • East = kelet ('rise'), as the Sun rises in the East
  • West = nyugat ('set'), as the Sun sets in the West

(Hungary is in the Northern hemisphere, so its vocabulary corresponds to the Sun's appearances there. – The above can be observed with the Latin word meridies, which means 'noon' and 'south' alike.)

Two words for "red"

There are two basic words for "red" in Hungarian: "piros" and "vörös" (variant: "veres"; compare with Estonian 'verev' or Finnish 'veres'). (They are basic in the sense that one is not a sub-type of the other, as the English "scarlet" is of "red".) The word "vörös" is related to "vér", meaning "blood". When they refer to an actual difference in colour (as on a colour chart), "vörös" usually refers to the deeper hue of red. While many languages have multiple names for this colour, Hungarian is special in having two distinct "basic" colour words for red.

However, the two words are also used independently of the above in collocations. "Piros" is taught to children first, as it is generally used to describe inanimate, artificial things, or things seen as cheerful or neutral, while "vörös" typically refers to animate or natural things (biological, geological, physical and astronomical objects), as well as serious or emotionally charged subjects.

When the rules outlined above are in contradiction, typical collocations usually prevail. In some cases where a typical collocation doesn't exist, the use of either of the two words may be equally adequate.

Examples:

  • Expressions where "red" typically translates to "piros": a red road sign, the red line of Budapest Metro, a holiday shown in red in the calendar, ruddy complexion, the red nose of a clown, some red flowers (those of a neutral nature, e.g. tulips), red peppers and paprika, red card suits (hearts and diamonds), red traffic lights, red light district, red stripes on a flag, etc.
  • Expressions where "red" typically translates to "vörös": red army, red wine, red carpet (for receiving important guests), red hair or beard, red lion (the mythical animal), the Red Cross, The Red and the Black, the Red Sea, redshift, red giant, red blood cells, red oak, some red flowers (those with passionate connotations, e.g. roses), red fox, names of ferric and other red minerals, red copper, rust, red phosphorus, the colour of blushing with anger or shame, etc.

Kinship terms

In Hungarian there exist separate words for brothers and sisters depending on relative age:
younger elder unspecified
relative age
brother öcs báty fivér or
fiútestvér
sister húg nővér nővér or
lánytestvér
unspecified
gender
kistestvér (nagytestvér) testvér

(There existed a separate word for "elder sister", néne, but it has become obsolete [except to mean "aunt" in some dialects] and has been replaced by the generic word for "sister".)

Besides, separate prefixes exist for up to the 6th ancestors and descendants (although there are ambiguities and dialectical differences affecting the prefixes for the 4th (and above) ancestors):

parent grandparent great-
grandparent
great-great-
grandparent
great-great-great-
grandparent
szülő nagyszülő dédszülő ükszülő szépszülő
(OR ük-ükszülő)
child grandchild great-
grandchild
great-great-
grandchild
great-great-great-
grandchild
gyer(m)ek unoka dédunoka ükunoka szépunoka
(OR ük-ükunoka)

Ősszülő or ószülő, as well as óunoka might be used for the great-great-great- great-grandparent or child, respectively.

On the other hand, no lexical items exist for "son" and "daughter", but the words for "boy" and "girl" are applied with possessive suffixes. Nevertheless, the terms are differentiated with different declension or lexemes:

boy/girl (his/her)
son/daughter
(his/her)
lover, partner
male fiú fia barátja
female lány lánya barátnője

Fia is only used in this, irregular possessive form; it has no nominative on its own. However, the word fiú can also take the regular suffix, in which case the resulting word (fiúja) will refer to a lover or partner (boyfriend), rather than a male offspring.

The word fiú (boy) is also often noted as an extreme example of the ability of the language to add suffixes to a word, by forming fiaiéi, adding vowel-form suffixes only, where the result is quite a frequently used word:

fiú boy
fia his/her son
fiai his/her sons
fiaié his/her sons' (singular object)
fiaiéi his/her sons' (plural object)

Extremely long words

  • Megszentségteleníthetetlenségeskedéseitekért

Partition to root and suffixes with explanations:
meg- verb prefix; in this case, it means "completed"
szent holy (the word root)
-ség like English "-ness", as in "holiness"
-t(e)len variant of "-tlen", noun suffix expressing the lack of something; like English "-less", as in "useless"
-ít constitutes a transitive verb from an adjective
-het expresses possibility; somewhat similar to the English auxiliaries "may" or "can"
-(e)tlen another variant of "-tlen"
-ség (see above)
‑es constitutes an adjective from a noun; like English "-y" as in "witty"
-ked attached to an adjective (e.g. "strong"), produces the verb "to pretend to be (strong)"
-és constitutes a noun from a verb; there are various ways this is done in English, e.g. "-ance" in "acceptance"
-eitek plural possessive suffix, second person plural (e.g. "apple" -> "your apples", where "your" refers to multiple people)
-ért approximately translates to "because of", or in this case simply "for"

Translation: "for your [plural] repeated pretending to be undesecratable"

The above word is often considered to be the longest word in Hungarian, although there exist longer words like:

  • Legeslegmegszentségteleníttethetetlenebbjeitekként

leg|es|leg|meg|szent|ség|telen|ít|tet|het|etlen|ebb|je|i|tek|ként
"like those of you that are the very least possible to get desecrated"

These words are not used in practice, but when spoken they are easily understood by natives. They were invented to show, in a somewhat facetious way, the ability of the language to form long words. They are not compound words – they are formed by adding a series of one and two-syllable suffixes (and a few prefixes) to a simple root ("szent", saint). There is virtually no limit for the length of words, but when too many suffixes are added, the meaning of the word becomes less clear, and the word becomes hard to understand, and will work like a riddle even for native speakers. An example:

  • Töredezettségmentesítőtleníttethetetlenségtelenítőtlenkedhetnétek tör - ed - ez - ett - ség - mentes - ít - ő - tlen - ít - tet - het - etlen - ség - telen - ítő - tlen - ked(ik) - het - né - tek

"you [plural] could constantly mention the lack [of a thing] that makes it impossible to make someone make something defragmenter-free"

See also: List of tongue-twisters#Hungarian.

Writing system

Before the year 1000, Hungarians had a different writing system. When Stephen I of Hungary established the Kingdom of Hungary, the old system was gradually discarded. However, although not used at all in everyday life, it is still known and practiced by some enthusiasts. For more information about this writing system, see Old Hungarian script.

Hungarian is written using an expanded Latin alphabet, and has a phonemic orthography, i.e. pronunciation can generally be predicted from the written language. In addition to the standard letters of the Latin alphabet, Hungarian uses several modified Latin characters to represent the additional vowel sounds of the language. These include letters with acute accents (á,é,í,ó,ú) to represent long vowels, and umlauts (ö and ü) and their long counterparts ő and ű to represent front vowels. Sometimes (usually as a result of a technical glitch on a computer) ô or õ is used for ő and û for ű. This is often due to the limitations of the Latin-1 / ISO-8859-1 code page. These letters are not part of the Hungarian language, and are considered misprints. Hungarian can be properly represented with the Latin-2 / ISO-8859-2 code page, but this code page is not always available. (Hungarian is the only language using both ő and ű.) Unicode includes them, and so they can be used on the Internet.

For a complete table of the pronunciation of the Hungarian alphabet, see X-SAMPA Magyar nyelvhez (in Hungarian, but the table is obvious), which transliterates Hungarian letters into IPA and X-SAMPA characters.

Additionally, the letter pairs <ny>, <ty>, and <gy> represent the palatal consonants /ɲ/, /c/, and /ɟ/ (a little like the "d+y" sounds in British "duke" or American "would you") - a bit like saying "d" with your tongue pointing to your upper palate.

Hungarian uses <s> for /ʃ/ and <sz> for /s/, which is the reverse of Polish usage. The letter is /ʒ/ and <cs> is /ʧ/. These digraphs are considered single letters in the alphabet. The letter is also a "single letter digraph", but is pronounced like /j/ (English <y>), and appears mostly in old words. The letters <dz> and <dzs> /ʤ/ are exotic remnants and are hard to find even in longer texts. Some examples still in common use are madzag ("string"), edzeni ("to train (athletically)") and dzsungel ("jungle").

Hungarian distinguishes between long and short vowels, with long vowels written with acutes. It also distinguishes between long and short consonants, with long consonants being doubled. For example, lenni ("to be"), hozzászólás ("comment"). The digraphs, when doubled, become trigraphs: +=, e.g. művésszel ("with an artist"). But when the digraph occurs at the end of a line, all of the letters are written out. For example ("with a bus"):

... busz-
szal...

When the first lexeme of a compound ends in a digraph and the second lexeme starts with the same digraph, both digraphs are written out: lány + nyak = lánynyak ("girl's neck").

Usually a trigraph is a double digraph, but there are a few exceptions: tizennyolc ("eighteen") is a concatenation of tizen + nyolc. There are doubling minimal pairs: tol ("push") vs. toll ("feather" or "pen").

While to English speakers they may seem unusual at first, once the new orthography and pronunciation are learned, written Hungarian is almost completely phonemic.

Order of words

Basic rule is that the order is from general to specific. This is a typical analytical approach and is used generally in Hungarian.

Name order

The Hungarian language uses the so-called eastern name order, in which the family name (general, deriving from the family) comes first and the given name (specific, relates to the person) comes last. However, as a rule, names are represented in the western name order when used in foreign languages. Thus for example Edward Teller, the Hungarian-born physicist, is known in Hungary as Teller Ede. Prior to the mid-20th century, given names were usually translated along with the name order; this is no longer as common. For example, the pianist uses András Schiff when abroad, not Andrew Schiff.

In modern usage, foreign names retain their order when used in Hungarian. Therefore:

  • Amikor Kiss János Los Angelesben volt, látta John Travoltát.

translates to

  • When János Kiss was in Los Angeles, he saw John Travolta.

Pre-20th-century foreign personalities have often had their names Hungarianized: Goethe János Farkas (rather than Johann Wolfgang Goethe). This usage sounds odd nowadays when only a few well-known personalities are referred to using their Hungarianized names, such as Verne Gyula (Jules Verne), Marx Károly (Karl Marx), etc. A number of native speakers disapprove of this usage; the names of certain religious personalities, however, are always Hungarianized by practically all speakers, such as Luther Márton (Martin Luther), Husz János (Jan Hus), Kálvin János (John Calvin), etc.

Dates

The Hungarian convention on dates is: 2000. január 1.

The order is big endian: 1., year (general) 2., month (more specific) 3., day (most specific)

Addresses

Although address formatting is increasingly being influenced by Indo-European conventions, traditional Hungarian style is:

1052 Budapest, Deák tér 1.

So the order is 1. postcode, 2., city (most general) 3., street (more specific) 4., house number (most specific)

Vocabulary examples

Note: The stress is always placed on the first syllable of each word. The remaining syllables all receive an equal, lesser stress. All syllables are pronounced clearly and evenly, even at the end of a sentence, unlike in English.

  • Hungarian (person, language): magyar [mɒɟɒr]
  • Hello!:
    • Formal, when addressing a stranger: "Good day!": Jó napot (kívánok)!
    • Informal, when addressing someone you know very well: Szia! [siɒ] (it sounds almost exactly like American colloquialism "See ya!")
  • Good-bye!: Viszontlátásra! (formal) (see above), Viszlát! [vislaːt] (semi-informal), Szia (informal: same stylistic remark as for "Hello!" )
  • Excuse me: Elnézést! [ɛlneːzeːʃt]
  • Please:
    • Kérem (szépen) (This literally means "I'm asking (it/you) nicely", as in German Danke schön, "I thank (you) nicely". See next for a more common form of the polite request.)
    • Legyen szíves! (literally: "Be (so) kind!")
  • I would like ____, please: Szeretnék ____ [sɛrɛtneːk] (this example illustrates the use of the conditional tense, as a common form of a polite request)
  • Sorry!: Bocsánat! [botʃaːnɒt]
  • Thank you: Köszönöm [køsønøm]
  • that/this: az [ɒz], ez [ɛz]
  • How much?: Mennyi? [mɛɲɲi]
  • How much does it cost?: Mennyibe kerül?
  • Yes: Igen [iɡɛn]
  • No: Nem [nɛm]
  • I don't understand: Nem értem
  • I don't know: Nem tudom
  • Where's the toilet?:
    • Hol van a vécé? (vécé/veːtseː is the Hungarian pronouncation of the English abbreviation of "Water Closet")
    • Hol van a mosdó? – more polite (and word-for-word) version
  • generic toast: Egészségünkre! [ɛɡeːʃʃeːgynkrɛ] (literally: "To our health!")
  • juice: gyümölcslé [ɟymøltʃleː]
  • water: víz [viːz]
  • wine: bor [bor]
  • beer: sör [ʃør]
  • tea: tea [tɛjɒ]
  • milk: tej [tɛj]
  • Do you speak English?: Tud(sz) angolul? Note that the fact of asking is only shown by the proper intonation: continually rising until the penultimate syllable, then falling for the last one.
  • I love you: Szeretlek [sɛrɛtlɛk]
  • Help!: Segítség! [ʃɛgiːtʃeːg]
  • It is needed: kell
  • I need to go: Mennem kell

Controversy over origins

Mainstream linguistics holds that Hungarian is part of the Uralic family of languages, related ultimately to languages such as Finnish and Estonian.

  • For many years (from 1869), it was a matter of dispute whether Hungarian was a Finno-Ugric/Uralic language, or was more closely related to the Turkic languages, a controversy known as the "Ugric-Turkish war", or whether indeed both the Uralic and the Turkic families formed part of a superfamily of "Ural-Altaic languages". Hungarians did absorb some Turkic influences during several centuries of co-habitation. For example, it appears that the Hungarians learned animal breeding techniques from the Turkic Chuvash, as a high proportion of words specific to agriculture and livestock are of Chuvash origin. There was also a strong Chuvash influence in burial customs. Furthermore, all Ugric languages, not just Hungarian, have Turkic loanwords related to horse riding. Nonetheless, the science of linguistics shows that the basic wordstock and morphological patterns of the Hungarian language are solidly based on a Uralic heritage.
  • A theory also well-known (still in dispute) is that the Hungarian language is a descendant of the Sumerian. Some linguists and historians (like Ida Bobula, Ferenc Badiny Jós, dr Tibor Baráth and others) had been working hard for decades and had published many detailed works (e.g. , ), and, purportedly, also there are some significant archaeological findings in this matter (like the Tartaria tablets). However mainstream linguists reject the Sumerian theory as pseudoscience.
  • Hungarian has often been claimed to be related to Hunnish, since Hungarian legends and histories show close ties between the two peoples; also, the name Hunor is preserved in legends and (along with a few Hunnic-origin names, such as Attila) is still used as a given name in Hungary. Many people share the belief that the Székelys, a Hungarian ethnic group living in Romania, are descended from the Huns. However, the link with Hunnish is uncertain, and even more so as most scientists consider the Hunnic language as being part of the Turkic language family.

There have been attempts, dismissed by mainstream linguists, to show that Hungarian is related to other languages including Hebrew, Egyptian, Etruscan, Basque, Persian, Pelasgian, Greek, Chinese, Sanskrit, English, Tibetan, Magar, Quechua, Armenian and at least 42 other Asian, European and even American languages. See Pseudoscientific language comparison.

Finno-Ugric words

mi meie, me (we) ők nemad, (they) ez/itt see (this here) az/ott too (there) ki? kes? (who) mi? mis? (we) nem mitte (no) egy üks (one) kettő/más/iker kaks (two) három kolm (three) négy neli (four) öt viis (five) ember isik (man) gyerek, gyermek, pudre, /fi /fia laps (kid) feleség (-né, nej) /némber naine (wife) férj abikaasa, (husband) anya/szülő ema (mother) apa/a(család)fő isa тетя tetya (father) fa puu (wood) bőr/héj nahk (skin) vér veri (blood) csont/tetem kont, luu (bone) haj juuksed (hair fej/kupa (head szem silm (eye) orr nina (nose) száj suu (mouth) fog hammas (Teeth)

Hun - Hungarian words

neilim: nyilam, neiliam: nyilaim (my arrows)

neilit: nyilad, neiliat: nyilaid (your arrows)

neilej: nyila, neiliaj: nyilai (his arrows)

neilinkh: nyilunk, neiliankh: nyilaink (our arrows)

neilitekh: nyilatok, neiliathakh: nyilaitok (your arrows)

neilekh: nyiluk, meiliakh: nyilaik (their arrows)

fej: phe(I), fé(K) (head

agy, agyvelő, csontvelő: ajge(I) (brain)

kéz: kezi(I) (arm)

kézfej: su(I) (hand)

szem: szöm(I), szüm(K) (eye)

arc: girze(I) (face)

orr, hegycsúcs: ore(I) (nose)

száj: szá(h)(I) (mouth)

nyelv: til(I), nyelv mint beszéd: lezu(I) (tongue)

szív: szerti(I) szeretet: szertild(I), szerelem: szirünild(I), szerelmes: szirüni(I), szeretetre méltó:szirünesi(I), szíves: szirünesi(I) (heart, love, love, in love)

talp: talba(I) (sole)

See also

Bibliography

Courses

  • Colloquial Hungarian - The complete course for beginners. Rounds, Carol H.; Sólyom, Erika (2002). London; New York: Routledge. ISBN 0-415-242584.

This book gives an introduction to the Hungarian language in 15 chapters. The dialogues are available on cassette or CDs.

  • Teach Yourself Hungarian - A complete course for beginners. Pontifex, Zsuzsa (1993). London: Hodder & Stoughton. Chicago: NTC/Contemporary Publishing. ISBN 0-340-56286-2.

This is a complete course in spoken and written Hungarian. The course consists of 21 chapters with dialogues, culture notes, grammar and exercises. The dialogues are available on cassette.

These course books were developed by the University of Debrecen Summer School program for teaching Hungarian to foreigners. The books are written completely in Hungarian. There is an accompanying 'dictionary' for each book with translations of the Hungarian vocabulary in English, German, and French.

  • "NTC's Hungarian and English Dictionary" by Magay and Kiss. ISBN 0-8442-4968-8 (You may be able to find a newer edition also. This one is 1996.)

Grammars

  • A practical Hungarian grammar (3rd, rev. ed.). Keresztes, László (1999). Debrecen: Debreceni Nyári Egyetem. ISBN 963-472-300-4.
  • Practical Hungarian grammar: [a compact guide to the basics of Hungarian grammar]. Törkenczy, Miklós (2002). Budapest: Corvina. ISBN 963-13-5131-9.
  • Hungarian verbs and essentials of grammar: a practical guide to the mastery of Hungarian (2nd ed.). Törkenczy, Miklós (1999). Budapest: Corvina; Lincolnwood, [Ill.]: Passport Books. ISBN 963-13-4778-8.
  • Hungarian: an essential grammar. Rounds, Carol (2001). London; New York: Routledge. ISBN 0-415-22612-0.
  • Hungarian: Descriptive grammar. Kenesei, István, Robert M. Vago, and Anna Fenyvesi (1998). London; New York: Routledge. ISBN 0-415-02139-1.
  • Hungarian Language Learning References (including the short reviews of three of the above books)
  • Noun Declension Tables - HUNGARIAN. Budapest: Pons Klett ISBN 9789639641044
  • Verb Conjugation Tables - HUNGARIAN. Budapest: Pons Klett ISBN 9789639641037

References

External links

Encyclopaedia Humana Hungarica

Dictionaries

Online translators

Online language courses

Search another word or see done deedon Dictionary | Thesaurus |Spanish
Copyright © 2014 Dictionary.com, LLC. All rights reserved.
  • Please Login or Sign Up to use the Recent Searches feature