Category Archives: Hindi

Judith Mirville on Language

I really enjoyed this piece. Those idiots at Badlinguistics are going to hate this post so much, but nevertheless, I think she is mostly correct here. This post is definitely Beyond Highbrow! Something to strive for, commenters! Reach for the skies!

Judith Mirville writes:

English has on one hand grown easier by shedding most of the heavy declension and conjugation-based Germanic grammar of old Anglo-Saxon and old Norman French it also derives fully from, but on the other hand it has grown into one of the most difficult languages of the earth due to the fact that to master it in a workable way, you have to work with roots coming from just too many linguistic horizons, each one having its own rules of combination.

You have more words of French origin in English than remain in French proper. For instance jeopardy and legerdemain are no longer understood in French, and poisonous is no longer correct. More words of Latin and Greek origin than allowed for in real Latin and Greek, apart from the simpler one-syllable, quite often more purely Saxon words that form more numerous prepositional idioms in the popular language than there are words formed likewise in German or Dutch, not counting a larger array than in other languages of purely exotic words having no common roots with any of the main ingredients of English.

The only other one language I know to be quite difficult for foreigners wanting to go beyond the cafeteria level for that very same reason is Hindi. Its grammar has retained only very few of the original complex Indo-European forms, and you cannot master its vocabulary without understanding that even though a few words of daily usage were originally Sanskrit or Pali, they have now undergone much transformation not necessarily for the simplest.

Yet most of the everyday vocabulary used in polite conversations is deformed Arabic imported through Persian, itself a compound language from Old Iranian and Semitic languages.

There is also a whole array of more recently learned terms artificially derived from Classical Sanskrit when it comes to science or Hinduism.

There is also quite a wide array of even more recently learned terms artificially derived from Classical Arabic when it comes to political science, economics, politics or Islam of course, which is an obligatory subject of conversation for all even those who combat it.

This is not counting an even wider array of words imported from English since the British era which is now widening with the advent of globalization. Each of those variegated language sources imposes its own ways of lexical derivation and quite often its grammatical forms.

Hindi, like English, seems somewhat easier than Sanskrit or Tamil as you begin, though it is never as easy as broken or basic English. But like higher-level English, you never, ever come close as a foreigner to master a working knowledge of it for universities or big enterprises.

In German (as well as in many Indian languages other than Hindi), by contrast, you have a much harder time mastering the grammatical machinery as you start, quite like a Mercedes engine, but once you do and you also master the root word combination system, you access very rapidly the highest realms of German philosophical thought.

I perfectly agree with you in stating that the idea put forth by many linguists that all languages are equal in terms of difficulty and ease of learning is a piece of utter fallacy and mendacity.

This is somewhat true only in the very specific context of automated learning of everyday language reflexes to be used without thinking in various situations, as if one were a spy working among a very distant people, and having learned to pronounce with the right accent most automated answers to daily practical situations like ordering toasts and coffee, paying a traffic ticket… one has also to pass more unnoticed in that environment than another person speaking a neighboring and similar dialect with less ease but more ability to express his thought.

This linguistic egalitarianism only works with people who will never bother to express anything they love to say but rather conceal what they know and camouflage it under nonsensical conversation of the kind that will never elicit any suspicion of unorthodoxy, as was the case in early Soviet Union.

And it comes to no surprise that such a linguistic theory came along together with Marxism. This theory can also work quite well in the context of enforced intellectual limitation by a ruling empire over all cultures to be stultified in the same way. But as soon as you are bothering to excel in a language and say everything you would love to say in your own or want to make serious intellectual research, this is simply untrue.

Some languages are really hard to learn, and some others quite easy, though the reasons may vary. Some languages are more difficult due to their lack of relationship with your native one, and some are quite difficult even to their own native speakers.

This PC view about languages just tells us about the limitation of all language they want to impose on us: prohibiting real self-expression and allowing only for a narrow range of practical commands. As they do when they say all races are equal and should mix with each other: what they tell is not the truth, but their aim instead is for the creation of a general stultified world citizen where all possible ancestral talents cancel out each other in favour of sheer mediocrity except for the cunning to make money by fraud and accepting bribes from the higher strata.

Anyway it won’t work: the most mongrelized White-African-Arabic new underclass they wanted to promote as model to be followed by all in France turned out to have lost all personal qualities and prejudices by race and culture mixing … except conspiracy-finding antisemitism as a natural federating factor as epitomized by Dieudonné. The result is that the new-fashioned intellectual Jewish elite of Paris are panicking, developing their own local version of neoconservative thought and telling the White Frenchmen to preserve their heritage from Africanization and mongrelization.

What I cannot stand though is the contrary point of view manifested by race realists such as Gedalia Braun that Negro languages are always more simple and primitive in structure and lacking in the power to express many concepts making civilization possible like metric graduation in the expression of distance in space and time and the notion of appointment and faithfulness as well as a vocabulary needing a dictionary to be relied upon and maintained.

I happen to be a passionate speaker of Haitian Creole of the most purely hillbilly kind as the language of my main love in life, and what Gedalia Braun says is 100% dead wrong even though Creole is supposed to be the zero ground in terms of general linguistics and mental development.

First of all, there is an elaborate tense system in Creole. It seems non-existent only relative to French verbs. Actually it works quite in the same way as English in terms of  morphology and auxiliaries though the shades of tense and aspect meaning are as elaborate as in Classical Spanish. It is much more refined and detailed in expression than the tense system of German or of Hebrew which is without any refinement in its modern form. And we are not even talking about the East Asian languages which are said to devoid of the notion of taste and actually more like what one caricatures as a Negro language.

Like English, and for the same reasons, Creole vocabulary is actually huge and of complex derivation, even though it seems easy to catch it when you begin as a traveling salesman, before long, you realize you will never be over with it.

You’ve got three main levels of language.

One that outwardly looks like simplified French but is combined very differently according to syntactic rules more like Semitic languages, possibly Aramaic, and of semantic rules more like Germanic languages. It is also very detailed, accurate and flexible as regards the expression of movement in space and time. A few engineers I know say it is seducing as an instrument for expressing equations.

The second level is the voodoo one, which works according to a different syntax copied from the Gbe language where the determinant comes before the determined as in German, not afterwards as in the first level, and is used for psychic manipulation purposes and power politics.

A third level of language is used for reasons of communication and compatibility with the surrounding modern sophisticated world and comprises all terms of Latin and Greek etymology present in either French, English and Spanish, generally with a rather French pronunciation but the same meaning as in English, and also a greater freedom in forming new terms by Pseudo-Latin derivation.

I don’t know anything about the Piraha language of Amazonia, but after having read a book by a pastor (Everett) who said he had witnessed the marvel of nonthinking people using it, and it had only three vowels, ten consonants, and no structured sentences, I can assure you this guy has been played with by those “primitives.” After all, as an American Evangelist missionary, he deserved to be shot by a poisoned arrow, but they defended themselves in a grander way by neutering his brain, maybe by the use of other less poisonous botanicals.

What that missionary says in a frantic, ecstatic mood is pure delusion.

First of all, there is a consecrated non-wordy, non-analytic, non-recursive way of expression most delicious to use whenever feasible in many languages closer to ours.

Portuguese is one of the best known examples of it.

Even though Portuguese is a very intricate and rich, complex language as regards its literary form proper, it possesses a register of expression that is very difficult to pick up. You have to develop extrasensory modes of communication to do it.

In this register, you exchange only one-word whispered sentences (like so pode) conveying each one a world of implications, making the conversation more like birds’ concert so to speak. Maybe the Portuguese Catholic Inquisition made that a matter of survival at some time, but its reputation for mortal totalitarian control has been grossly exaggerated compared to other control-freaks like the Judaeo-Anglo-Saxon PC crowd.

Everett has remained in the same kind of racist outlook with direction only reversed. Actually, the Pirahas he has met with have always known much more about his culture and his world, together with many other ones that have been threatening them into extinction for centuries, and which they have circumvented through manipulation so far, than he has about theirs, even after all he thinks he has discovered. I suspect the Pirahas to be a very cunning and not so charming and benevolent crowd, though capable of huge good practical jokes: not at all the last castaways from Eden that Everett still imagines as a former Evangelical.

There is certainly a huge higher initiation level of language the Pirahas are dead intent on reserving to themselves, which as high in left brain content as KGB Russian, the same level as in Portuguese, and my beloved Creole. Haitians even used to have computer-like programming languages long before computers, except that they were used to program humans made into zombies, and the purpose of them was always evil.

Arabic, among the languages of worldwide use, is one of the most difficult technically, not only because of its non-relationship with any roots we know in our own languages or its very heavy and irregular morphology as regards plurals, conjugations, declensions and its convoluted syntax, but also because very simple notions in most other languages even in supposedly closely-related Hebrew never can be said in clear simple terms in Arabic and need a cumbersome grammatical apparatus to be conveyed.

To express the concept of doing again or re-doing something, you have to fully conjugate the two verbs re- and do (prepositions are conjugated too, with as many special rules as with verbs), you cannot add something like un- or de- to express the undoing of something.

Instead you have to use a full clause like I am undoing the attachment of my shirt instead of I am untying it. You cannot say I have done it already, instead you have to say something like It is already overtaken by my doing it. You generally don’t say I must do it (even though you could in theory), instead you more commonly say There is no alternative for me apart from doing so.

One thing I like about Arabic though is its closure towards foreign admixture and the difficulty for foreign words to get naturalized, with the result that the semantic universe is simpler than elsewhere and more coherent.

The most difficult aspect though is that you cannot form compound terms and verbs the way you do in English and Romance languages by using suffixes and prefixes, especially when as a stranger or a beginner you are short of the exact term and would use a synonymous compound word instead. Even the negation of adjectives is not guaranteed, and you have to learn the contrary ones which have independent roots.

One thing remarkable about Arabic is the utmost difficulty of expressing in it the idea of excess or of extremism as being an undesirable thing, and conversely, of moderation as being a virtue. The word too or too much simply doesn’t exist. Phrase books and Google translation recommend to use the word very (jiddan) instead, but it has nearly always a laudatory connotation, and if you insist in using it for meaning too much you are spotted as a clumsily-speaking foreigner.

The problem is that practically all comparatives and superlatives that are used to render the idea of relative excess to a situation, like a truck too high for a tunnel to pass through, are also by themselves as elatives having an admirative value. When you say akbar for instance, it is very big or bigger than expected, but it can never really be too big. It is always something like “Wow my Gosh, how it’s big!” Even apart from the worldwide known religious and terroristic use of Allahu Akbar proper, it is just too big eventually for the sum of money it would cost or some other accidental impediment like a ceiling.

In theory, in very Classical (though non-Koranic) Arabic, you could also use a difficult conjugated verb in a serial clause for expressing the simple adverb too much (the verb ifrat:a, meaning overdoing) as is the case with most simple English adverbs, but  that would sound as pedantic, unnatural and unusual as Shakespearean “multitudinously” (except as verbal nouns to form scientific compound terms used in universities only) and make everybody around laugh, even among religious speakers of Classical Arabic only.

The word “moderate” is generally recommended to be translated in journalistic lingo as mutaäddil.

But if you leave the Western-style university class for the university cafeteria and say Ana muslimun mutaäddil (I am a moderate Muslim), your colleague from a non-Western culture-related subject will understand something completely different.

He will know that your appetites are well moderated by your faith in Islam, that you have renounced all alcohol, you no longer smoke, you skip one meal out of two and fast for the whole Ramadan, you never indulge in erotic or profane literature and try live a spartan life in order to spare money for the Hajj, which things are not a promise of tolerant conduct towards non-believers.

All good translators into Arabic will tell you of the challenge to render such an expression as too much or of the general concept that an extremist point of view (mutat:arrif) might be condemnable.

Leave a comment

Filed under Afroasiatic, Applied, Arabic, Balto-Slavic-Germanic, English language, French, German, Germanic, Hindi, Indic, Indo-European, Indo-Hittite, Indo-Iranian, Indo-Irano-Armenian, Indo-Irano-Armeno-Hellenic, Italic, Italo-Celtic-Tocharian, Language Families, Language Learning, Linguistics, Portuguese, Romance, Sanskrit, Semitic

A Look at the Indo-Aryan Languages

From here.

This post will focus on how hard it is to learn the various Indo-Aryan languages if you are an English language speaker. We will look at Kashmiri, Hindi, Bengali, Punjabi, Sinhala and Sanskrit.




Ind0-Ayran languages like Kashmiri, Hindi and especially Sanskrit are quite hard, and Sanskrit is legendary for its extreme complexity. Sanskrit grammar is very complicated. There are 8 cases. However, Bengali is said to be one of the easiest Indic languages to learn.

Central zone
Western Hindi

The Hindi script is quite opaque to Westerners, some of whom say that Chinese script is easier. You speak one way if you are talking to a man or a woman, and you also need to take into account whether you as speaker are male or female. Gender is also as prominent as in Spanish; you have to remember whether any given noun is masculine or feminine.Hindi is definitely an IE language by its rich system of gender, case and number inflection.

The most difficult aspects of Hindi are the pronunciation and the case system.

The distinction between aspirated/unaspirated and alveolar/retroflex consonants is hard for many to make. There is a four-way distinction ion the t and d sounds with aspirated/unaspirated dental and aspirated/unaspirated retroflex t‘s and d‘s. The are three different r sounds – one that sounds like the English r and two retroflex r‘s that are quite hard to make or even distinguish, especially at the end of a word. Hindi also has nasalized vowels.

If you come from a language that has case, Hindi’s case system will not be overly difficult.

In addition, there is a completely separate word for each number from 1-100, which seems unnecessarily complicated.

However, Hindi has a number of cognates with English. I am not sure if they are Indic loans into English or they share a common root going back to proto-Indo-European (PIE).

loot plunder/destroy, English loot.
season/weather, English equivalent is monsoon
storm, English equiv. typhoon
d – something tied around the waist, English equiv. cummerbund
– literally bad name, means bad reputation. These are both cognates to the English words bad and name.
bangalaahouse, English equiv. bungalow
priest, English equiv. pundit

Hindi is rated 3.5, harder than average to learn.


Bengali is similar to Hindi, but it lacks grammatical gender, and that fact alone is said to make it much easier to learn. Bengali speak tend to make stereotypical gender errors when speaking in Hindi. Nevertheless, it uses the Sanskrit alphabet, and that alone makes it hard to read and write.
Bengali is rated 3.5, harder than average to learn.


Punjabi is probably harder than any other Indic language in terms of phonology because it uses tones. It’s like Hindi with tones.
Punjabi is rated 4.5, extremely difficult.


Sinhala is also difficult.
Sinhala is rated 4, very difficult.


Sanskrit is legendary for its difficulty. It has script that goes on for long sequences in which many small individual words may be buried. You have to take apart the sequences to find the small words. However, the words are further masked by tone sandhi running everything together. Once you tease the sandhi apart, you have to deal with hundreds of compound characters in the script. Once you do those two things, you are left with eight cases, nine declensions, dual number and other fun things.

Even native speakers tend to make grammatical mistakes are admit that parts of the grammar are fiendishly difficult. There are many grammatical features that are rarely or never found in any other language. Noun declension is based on the letter than the noun ends in, for instance, nouns that end in a, e or u all decline differently. There are three genders for nouns, and those all decline differently also. Each noun has eight cases and three numbers (singular, dual and plural) so there are 24 different forms for each noun. Counting the different combinations of endings and genders (all subsumed into a sort of noun class system) there are 20 different “noun classes.”

Combining the “noun classes” with the three genders, you end up with 1,440 different regular forms that nouns can take. To make matters worse, some of the cases have different forms themselves. And there are some exceptions to these rules. The I and you pronouns decline differently, but pronouns are simple compared to nouns.

For the verbs, each verb had exist in 10 different forms of tense or mood (one from Vedic Sanskrit is no longer used). There are six tenses and four moods. The six tenses are: one present tense, 2 future tenses and 3 past tenses. The moods are: imperative, dubitive (expresses uncertainty), optative (expresses hope or offers a benediction) and a form that expresses the concept if only, then… There are two different conjugations based on who is the beneficiary of the action, you or others. There are ten different classes of verbs, each of which conjugates differently. Additionally, each verb has a different form in the singular, dual and plural and in the 1st, 2nd and 3rd persons.

Once you get past all of that, you are ready to take on the really difficult parts of the language, participles, noun derivatives and agglutination, each of which is far more complicated than the above.

Nevertheless, the language is so mathematically precise and regular that some have said it is a perfect language for computer programming. There may not be a single irregularity in the whole language.

Sanskrit is rated 5, the most difficult of all.

Kashmiri is rated 4, very difficult.


Filed under Applied, Hindi, Indic, Indo-European, Indo-Hittite, Indo-Iranian, Indo-Irano-Armenian, Indo-Irano-Armeno-Hellenic, Kashmiri, Language Families, Language Learning, Linguistics, Sanskrit, Sinhala

Osama bin Laden on Top of the World

Repost from the old site. Dated but fascinating stuff.

Where is Osama bin Laden, and where has he been hiding ever since he fled Tora Bora in December 2001? There are many theories about where he hides. One theory is that he has been and is hiding in Pakistan.

I do not speak of the region encompassing North and South Waziristan, where bin Laden and Ayman Al Zawahiri have long been rumored to be. Nor do we include Peshawar, which seems too busy to hide him, although he probably stayed there in January and February 2002, being carefully hidden by members of Pakistan’s major fundamentalist parties who now rule the area.

Nor do we even speak here of the Bajaur Tribal Agency or the adjacent Afghan provinces of Kunar and Nuristan, where this blog has long suspected that bin Laden may be, and where he has definitely been hiding off and on since then. For instance, Pakistani President Musharaff said in mid-September 2006 that Bin Laden is hiding somewhere in Bajaur, Kunar and Nuristan.

On September 28, 2006, Pakistani sources said that a new Ayman al-Zawahiri tape was circulating but had not yet been released. They said it had been recorded around the border between Bajaur and Kunar. On October 31, 2006, Pakistani intelligence said they think that they have Zawahiri “boxed in” in a 40-mile square area bordered by the Khalozai Valley in Bajaur and the village of Pashat in Kunar, and they hope to capture him in a few months.

There is also evidence that bin Laden is in that area. For instance, Osama reportedly attended a wedding of a daughter in or near Bari Kot in Kunar Province in 2002. In September 2003, according to a Newsweek article on September 8, bin Laden was said to be hiding north of the Pech River Valley in Kunar.

The source in the article was an interview with an Afghan whose daughter was married to an Algerian member of Bin Laden’s “praetorian guard”. The Algerian occasionally came down from his hideout to visit his wife.

Bajaur is thought to be Al Qaeda’s “winter hideout”. Zawahiri has surely been in Bajaur, first at Damadola, where he was nearly killed by an airstrike that missed him by mere hours, and in Chinagai, where a rumored Zawahiri visit resulted in a recent US-Pakistani strike that killed 80 madrassa students.

Afghans captured at a safe house in Bajaur in May 2005 reported that the house was regularly visited in February and March 2005 by a masked man with a large bodyguard contingent whom they suspected to be Zawahiri. I have been unable to pinpoint the town in Bajaur where the safe house was.

We are not talking about those areas in this post.

Instead, we are talking about a region in Pakistan’s Northwest Frontier Province ranging from the Dir District north of Bajaur and adjacent to northern Kunar Province in Afghanistan, up into the Chitral District through Lowarai Pass along the Chitral Valley, past Drosh, and up to and beyond Chitral, the old hippie and trekker heaven across from Nuristan Province in Afghanistan.

From the Chitral Valley, we move east of Drosh and back into the Dir District to the Kumrat Valley in Dir Kohistan, part of a larger area called Kohistan. From Chitral, we move north up to Tirich Mir, a 25,000 ft. peak that soars to the skies. From there, we move west to Salim Shah, near the Dorah Pass on the border of Pakistan and the Afghan provinces of Nuristan and Badakhsan.

From that point, we move east along the Wakhan Corridor of Afghanistan first to Darkot, near 15,000 foot Darkot Pass, and then further east, out of the Chitral District and the NWFP into the Ghizar District in Kashmir, to an extremely remote town called Chillinji, near the Chillinji Pass, a 17,358 foot pass near where Afghanistan, Pakistan and China all come together.

Some places around here, especially north, northwest and west of Chillinji have just recently been explored by modern explorers, who were shocked to find almost no maps or trip reports available when they planned their trips.

From Chillinji, we move a ways east, out of the Hindu Kush and into the Karokoram Mountains, into the Gilgit District to just south of the Chinese border in the Hunza.

Along the way, we will check out some breathtaking photos of the region.

This blog has already reported on this possibility in a prior post, which focused somewhat on Chitral but also investigated reports that bin Laden was seen at an Al Qaeda training camp in the Minteke Pass region on the border of Pakistan and China in 2002. That post included spectacular photos of Chitral and many other photos of the Karokoram Mountains and the Hunza area, which encompasses the Mingteke region.

Lending possible credence to theory that Bin Laden is in the high mountains of the Hindu Kush, at least in the Summer, when most of the reports have placed him there, is a theory that the CIA is operating on the seasonal movements of Al Qaeda’s top leadership.

The theory is that the leadership moves between a number of safe houses at lower elevations in the winter (in areas that are pretty much snowed in) and then moves to caves at very high elevations in the Summer when the lower elevations they had been staying in become accessible.

If you think about it, the best place to hide in the region in the warmer months is as high up as possible. A recent ABC news investigative report video segment placed bin Laden between the Khyber Pass and Chitral along the Afghan-Pakistan border. That video also noted the seasonal movements described above (works only in Internet Explorer – shame on ABC!)

This very nice and easy-to-read map is a good guide to at least part of this post. The red circle is the approximate location of Damadola , where a US Predator tried to kill Ayman Al-Zawahiri (Al Qaeda’s Number 2 man) when he was visiting in-laws on January 13, 2006. (Zawahiri married a woman from the Damadola area in 2001.) The missile only missed the Al Qaeda leader by hours and killed Al Qaeda’s top explosive expert.

Abu Marwan Al-Suri, Al Qaeda’s money man, was killed on April 20, 2006 between Khar (not on the map), capital of Bajaur, and the Afghan border. That would also be in about the location of the red circle on the map.

Dargai on the map is where the Pakistani Taliban suicide bomber attacked a Pakistani base recently, killing 45 recruits, in a revenge attack for the US Predator attack on the Al Qaeda training camp/madrassa in Chinagai (not shown, but located by the bend in the road east of Mian Kalai on the map) on October 30, 2006, killing 80 madrassa students. That attack may also have been based on a rumored visit by Zawahiri.

Abu Faraj Al-Libbi, the so-called Number 3 man in Al Qaeda, was arrested in May 2005 in Mardan on the map. Dir, Drosh and Chitral on the map are discussed in this post. The town of Tal on the map is located in the Kumrat Valley, discussed in the post.

Bin Laden and Zawahiri are also often said to be hiding in the Afghan province of Nuristan around Kamdesh on the map. Bin Laden appeared at a wedding for a daughter in 2002 in Barikot on the map in the Afghan Province of Kunar in 2002.

Tirich Mir, Darkot and Chillinji Pass are off the map to the north and northeast.

We begin our journey in Dir, where Frontline (a US Public Broadcasting Service, or PBS, program) reported that Osama bin Laden was seen only three weeks before they visited on August 30, 2002 to make the special In Search of Bin Laden. During Frontline’s visit to Dir, they described it as totally hostile. Bin Laden had also been seen in Chitral just weeks prior to the Dir sighting. At the time, Zawahiri was thought to be operating out of Chitral.

Later, in March 2003, we have a report from the Iranian News Agency that an FBI team (actually CIA) was in Chitral in order to intercept a planned meeting between bin Laden and Gulbuddin Hekmatyar, the leader of Hizb-Islami, an Afghan radical Islamist group, who had just thrown his hat in with Al Qaeda and the Taliban.

In truth, there were extensive searches in Chitral in February-March 2003 involving both CIA and Pakistani forces. The two-person CIA team (male and female) cruised the district in an SUV, but found nothing. Parties were reportedly coming across the border from Nuristan for the meeting.

Abu Khabaib, an Arab explosives master, has been spotted several times in the Chitral Mountains. He has worked with Sheik Ahmed Salim, an Al Qaeda operative who has given money and fighters to Lashakr-e-Jhangvi, a Pakistani extremist organization that is primarily dedicated to waging horrible terrorist attacks (often large-scale attacks) on Pakistan’s Shia population.

Al Qaeda released a videotape in September 2003 showing bin Laden and Zawahiri picking their way down a very steep and high-elevation forested mountain. The trees in the videotape are reportedly peculiar to the Chitral region.

The Chitral District has a population that is 65% Sunni, 35% Shia Muslim and a tiny percentage of non-Muslim Kalash, who practice a pre-Islamic native religion with overtones of animism.

This obscure blog feels that bin Laden is hiding between Dir and Chitral in the Chitral Valley, which would put him around Drosh on the map. A description of the Frontline crew’s visit to Drosh on September 1, 2002, with a nice photo, is here.

Moving on up the Chitral Valley past Drosh, we come to Chitral, where hippies migrated in the 1960’s and 1970’s for its legendary hashish. Even back then, it was very difficult to get in, as all entry to foreigners was banned, and you could only get in with an invite from a resident.

When Frontline traveled through the Lowarai Pass between Dir and Drosh, they looked in the guestbook and noted that only two Americans had been there since 9-11, a period of almost a year. The soldier manning the post joked and said that one of the two Americans was a spy.

Stomach-churning switchbacks on the road to Lowarai Pass. This terrible road is one of the best roads in this part of Pakistan. The Phalura people live here. Their language is related to the Kohistanis 40 miles to the west; the Phalura apparently moved to this area from Kohistan several centuries ago for unknown reasons. Bin Laden and Zawahiri were apparently in this area in July and August 2002. Are they there now?

The people around the pass speak a Dardic language with 8,600 speakers called Phalura, that is related to Shina (Shina is described at the end of the post). In some villages, the language appears to be dying in favor of the more widely spoken Khowar but in others it is still vibrant .

The people here are Sunnis. A linguist in Sweden, Hendrik Lilgendren, is working on a grammar of the language, a much-needed addition to our knowledge of these poorly documented languages of Northern Pakistan.

Chitral is now a trekking capital for intrepid visitors all over the world, but it is still a great place to hide. There are tourists in the summertime, but in the winter, Lowarai Pass, its window to the world, is generally snowed in. Air access to Chitral is meager at best, and the electricity and Internet go out all the time. In 2006, perhaps ominously, Pakistan closed the entire Chitral region to tourists due to “security reasons”.

The local governor has reportedly forbidden intelligence officers from searching the area and local politicians have been protesting, perhaps a bit too much, that bin Laden is nowhere in the area. Fundamentalist Islam is big even in Chitral, where the local Islamist party in summer 2006 protested women having their own stalls in the market.

In late August 2006, Al Qaeda authority Peter Bergen said US intelligence was starting to focus on Chitral as a possible hideout for bin Laden. This conclusion was based on an analysis of the trees in Al Qaeda videotapes and the length of time it seems to take to deliver Al Qaeda tapes after they have been made (about 3 weeks).

Another interview with Bergen for a CNN program called “In the Footsteps of Bin Laden” aired in September 2006 elaborates along the same lines, and says that Chitral is considered “inaccessible” and that US forces are banned from the area at any rate.

Note this recent article from September 2006, though, that notes that since January 2006, a special US bin Laden task force called Task Force 145 is now allowed to go into Pakistan without permission to hunt bin Laden.

Chitral, where as recently as August 2006, US intelligence felt that Osama bin Laden was hiding. A CIA team rented a home here to search for him until they were chased out in Spring 2006. Some think he was living a quiet, paranoid life here in a home with only a few other people. He would hardly ever go out, and his visitors would be very tightly regulated. He may have spent his day listening to radio, watching the news and surfing the net, since he is a news junkie.

A US intelligence official was quoted as saying that bin Laden was thought to be living not in a cave but in a house in Chitral, possibly with a family, and with a small bodyguard team of maybe only a couple of guards. Bergen felt that bin Laden spendt his days on the Internet, watching CNN and listening to BBC radio. He may access the net through a device called an HF modem that connects users to the net via radio waves.

Pakistani officials protest that Chitralis do not like bin Laden due to ethnic and religious differences, so it is unlikely that he is in the area. A tiny CIA (reports say it was FBI, but CIA here is always referred to as FBI, in order not to inflame Pakistani sensibilities) team rented a home in Chitral in Winter 2005 under the cover of being investors trying to bring business to the area.

By Spring 2006, local politicians found out about them, and were raising a fuss about them after their location was revealed by a local politician, but by that time, they were already gone.

There is another report, this one from June 17, 2005, from ABC’s Brian Ross. In this case, a group of Arabs came down from Nuristan to a small market town in the Chitral District and loaded up on large quantities of supplies, including wheat and flour. Then they headed back to Afghanistan. I do not know what town in Chitral they came to. It seems probable that this group was connected to the Al Qaeda leadership.

People in Chitral mostly speak Khowar, a Dardic language with 240,000 speakers that is distantly related to Hindu and Persian and more closely to Punjabi, Kashmiri and Sindhi. You can see the tremendous differences between it and Pashto, Hindi, and even Nuristani in this lexical chart. Note the occasional cognate though, such as “eye”.

About 15-20 % of speakers are literate in the language. Nevertheless, Khowar, I am happy to say, is being well-developed as a literary language.

Adding weight to the theory that bin Laden may be or may have been in Chitral is a recent report from ABC News that more foreigners have been seen lately in the Waziristan and Bajaur Agencies of FATA, and in the Dir and Chitral Districts of NWFP. In all of these areas (including Chitral), the Taliban have been openly recruiting for jihad in Afghanistan.

The heightened presence of foreigners and open recruiting is apparently due to recent peace deals between the Pakistani government and jihadis in North and South Waziristan that essentially handed over power there to Al Qaeda and the Taliban.

Moving north of Chitral, this article from around the same time, late August 2006, says bin Laden was thought to be hiding north of Chitral, around the towering Tirich Mir Mountain, a soaring peak that rises to 25,289 feet. No source was given, but Western intelligence agencies are usually the sources of these tidbits.

Tirich Mir Mountain, where in August 2006, US intelligence sources suspected Osama bin Laden may be hiding.

A nice webpage on this mountain, with some spectacular pictures, is here.

The Chitral River near Tirch Mir, north of Chitral. This river cuts a wide swath through the canyons here. Good luck getting across it.

You were wondering how you get across the Chitral River in this area? Like this. As you can see, motor vehicles cannot cross here and typically pack animals cannot either.

There are plenty of caves around there too, sympathetic tribesmen, a ban on US forces operating in the area, and few to no roads. Another great place to hide.

Goats on the road near Tirich Mir, north of Chitral. Is that a paved road? Amazing.

On a map, Tirich Mir can be seen in the far northwest corner of Pakistan, north of Chitral and south and west of the border with Afghanistan.

Kuragh Ridge, near Tirich Mir north of Chitral. An intelligence report in August 2006 placed Osama bin Laden here.

A road in the Yarkuhn Valley, about 1/2 way between Tirich Mir and Darkot. This is the best road in the area!

The area circled shows the location of various areas discussed in this post, including Chitral, Tirich Mir, Drosh, Dir, Salim Shah, Dorah Pass, Garam Chasma, the Kuragh Ridge and the Kumrat Valley.

Moving east, a report by ABC’s Brian Ross from May 24, 2006 (6 months ago) said that Pakistani officials had received reports that Al Qaeda’s top leadership was in the Kumrat Valley of Dir Kohistan, 40 miles inside Pakistan. They had reportedly moved there from Chitral.

Kohistan, where the Al Qaeda leadership was thought to have moved to from Chitral in summer 2006. The trees and terrain here do resemble that seen on a videotape of bin Laden and Zawahiri in 2003, I must admit. This photo also looks very much like Nuristan Province in Afghanistan, where the leadership has also long been rumored to be hiding. It also reminds me of Idaho in the US Northwest.

The Pakistanis felt that the bin Laden’s entourage had recently moved down from the high peaks near the Afghan-Pakistan border to Kohistan.

Stunning Kohistan. This photo looks like the Caucasus Mountains on Russia’s southern border. What a great place to hide! Why not hang a Welcome Al Qaeda sign?

The Kumrat Valley is identified on the map above by the town of Tal, which is located in the valley.

A nice web page on Kohistan is here . Kohistan (population 2 million ) is one of the most isolated and deprived areas in Pakistan, and the illiteracy rate ranges from 94-98%!

A dwelling, or series of dwellings, on a terraced field in Kohistan. Crops do not grow well in this region at all, with poor weather and rocky soils being major factors.

There are 5 doctors per million population in Kohistan – or 1 doctor for every 200,000 people.

Harsh winters in stunning Kohistan add to the isolation of this region.

The people in the Kumrat Valley speak a language called Kalami, which has 40,000 speakers. In this area, the Thal dialect of Kalami is spoken. This dialect has around 90% similarity with most of the other Kalami dialects. Most men also speak Pashto.

This link has an excellent description of the people and language of this region (pdf). Since the creation of an alphabet in 1995, Kalami is now a written language with an Indo-Arabic script similar to Urdu. Three books have now been published in Kalami by the Kalam Cultural Society. Links to two of those books are here and here. Some stunning photos of the Kalam area can be seen here.

The Kalamis are similar in many ways to the Kohistanis of Afghanistan. They are all strict Hanafite Sunni Muslims.

As you can see, the natives here in Kohistan are not particularly friendly. “Who are you looking for? Osama who? Never seen him.” Afghan Nuristanis often have similar hostile expressions towards visitors.

As with the Kohistanis in Afghanistan, almost all men are armed at all times in order to defend themselves against their numerous enemies, since most of them are engaged in continuous feuds with other Kalamis, usually revolving around women and honor.

A Pakistani official recently stated that Kohistan is the most lawless area of all Pakistan, and Dir Kohistan was the most lawless part of Kohistan. Feuds kill about one person every two weeks in the Tal Region.

The strict Sunni Muslims here in are all armed, all the time, and ready to fight. That and they don’t like strangers, especially nosy infidels. For bin Laden, Kohistan is paradise.

Kohistani women are treated terribly in Kohistan, as they are in Afghanistan. Kohistani women are treated even worse than Pashtun women, which is pretty bad!

This region is extremely inaccessible. There are only two roads in the entirety of Pakistani Kohistan, and those are pretty bad.

A typical “road” in Kohistan. Inaccessible by any motor vehicles; you need to access places like this either on foot or by donkey. Nuristan in Afghanistan has similar terrain and a similar “road” system. The first road in Nuristan was reportedly built in 1979.

Most travel here is on foot or by mule on narrow rocky trails.

Bridges are often narrow hanging bridges over raging, torrential rivers tearing through soaring canyons. The bridges are so narrow and rickety that often pack animals will refuse to cross them, rendering much of the area inaccessible even by mule!

That is called a bridge in Kohistan – a log over a raging torrent of a river plunging down a skyscraper canyon. If you hide here, no motorized vehicle can get to you. Search teams must come by mule or on foot and you will be warned far in advance.

These same conditions regarding roads and bridges also apply to Nuristan, and similarly limit access to that Afghan province and make it an excellent place to hide.

That’s barely even a trail. This is what passes for a road here in Kohistan. An enemy army here is confined to mules or boots. This is the terrain US forces have to deal with in Kunar and Nuristan in Afghanistan. Have fun!

As you can see, this is an excellent place to hide! US forces are not allowed here and Pakistani forces cannot access most of the region via motorized vehicles.

A typical hanging bridge in Kohistan. Nuristan in Afghanistan has many similar bridges that often look remarkably like this one.

Pakistani troops would be limited to pack animals, and in many cases, they would just have to abandon the animals and go on foot.

Often, pack animals simply refuse to cross these Kohistan foot bridges over raging alpine rivers. Surely, anything with four wheels is useless here. The bridges and scenery in Afghanistan’s Nuristan are quite similar.

The people here are notoriously hard to govern, and early observers reported that this area was inhabited by “ anarchists” who opposed all outside authority. The terrain here is spectacular, with plunging canyons, soaring peaks, raging rivers, lush meadows and deep evergreen forests.

Lush forests and soaring peaks in this photo of Kohistan look like the Sierra Nevada near my home. The forests here are still intact, unlike those in much of the region.

The forests here are vast and the deforestation level is still low, unlike Nuristan and Kunar.

Spectacular scenery in Kohistan as a smaller tributary meets a larger river amidst soaring, impenetrable forests. This could be Alaska, Western Canada (British Colombia) or Norway.

There is also a very healthy wildlife population in Kohistan, including rare animals. Large carnivores, often rare elsewhere in the region, are abundant. Brown and Black Bears, leopards and snow leopards and wolves roam the wild forests.

Kohistan is a paradise for wildlife, even rare predatory mammals that are barely holding their own elsewhere. Can you imagine Al Qaeda’s most wanted hiding here? This photo looks like the Rocky Mountains in Canada.

What few roads and trails existed prior to 1992 have now been wiped out after terrible floods raged through Kohistan in that year, destroying roads, trails, bridges and structures.

There are a few roads in Kohistan, if you can call that a road! This is transportation at its finest in this region.

Most of the area has still not yet been rebuilt.

Floods raged through these canyons in Kohistan 14 years ago, devastating an already deprived area and isolating it even further.

The people grow a few crops here, but only 4% of the land is cultivated and most crops do not grow well.

Rain in Kohistan, with snow on the ground. The climate here is extremely wet, with lots of snow in winters and plenty of rain the rest of the year.

They also increasingly engage in forestry and especially the gathering of medicinal herbs for sale, of which there are around 70 types growing here.

The gorgeous forests of Kohistan. By examining the trees in an Al Qaeda video from 2003, the CIA thinks it is pinning down where bin Laden is located. This photo could be of the Black Forest in Germany, the US’ Washington state and Canada’s province of British Colombia in the Pacific Northwest, Chile’s Patagonia, or Russia’s endless Siberian forests.

For the most part, residents subsist on animal husbandry. They obtain milk, meat, cheese, wool and hides from these animals.

Residents in Kohistan graze animals on the gorgeous alpine meadows here. This looks like the Alps.

In recent years, quite a bit of tourism has come to the Kalam area, but I don’t know how they are getting around unless they are trekkers.

Moving even further to the north and off the map above, in this fascinating Christian Science Monitor article by Scott Baldouf on August 9, 2002, Al Qaeda Massing For New Fight, Ayman Zawahiri was said to be holding sway at a new base constructed by Al Qaeda in the town of Shah Silim (map).

A village near Garam Chasma, which is close to Shah Salim, where Al Qaeda reportedly had a large base in 2002. Look at those craggy, forbidding peaks.

This obscure report notes that in August 2002, echoing Baldouf’s article, bin Laden and Zawahiri traveled to Dir and Shah Salim to solidify their alliance with Gulbuddin Hekmatyar, head of the Hezb-i-Islami. It also lists in fascinating detail all of bin Laden’s known movements and activities from November 2001 to August 2004.

The road along the Lutko River from Chitral to the Dorah Pass is actually accessible by jeep. This is near Shah Salim.

Shah Salim is located about 30 miles northwest of Chitral on the Afghan-Pakistan border near where Pakistan meets the Afghan provinces of Badakhshan and Nuristan.

Spectacular footage and forbidding terrain near Garam Chasma, close to the Shah Salim Al Qaeda base. The people here are speak a language called Yidga.

On a map, it is located between the town of Darband and 14,940 foot Dorah Pass, only a few miles from the Afghan border. This pass played an important role in the war, as this was an important supply route form Pakistan to Afghanistan that the Soviets were never able to shut off.

The Afghan side of Dorah Pass, the famous route used by the mujaheddin in the Soviet War. The people in this area used to be mostly Munjis, but most of the Munjis left to Pakistan with the war. It’s not known if they have returned or not yet.

This is an extremely wild, remote and inaccessible region. There are hardly any roads and much of the access is by foot or horse. In 2001, the area was experiencing a terrible humanitarian crisis. Most of the people here speak Yidga (6,145 speakers) and are Ismaili Shia Muslims. Yidga is a Pamir language. The Pamir languages are Indo-Iranian languages related to Persian and Pashto, but fairly distantly.

Across the border in Afghanistan, people speak another Pamir language called Munji that is somewhat closely related to Yidga (about as close as English and Dutch). Most Munjis fled Afghanistan and settled in Pakistan during the Soviet War.

A fascinating Australian Medical Journal article in 2002 reported on the experience of a doctor, Robert Simpson, working with MSF (Doctors Without Borders) in the region in 1999, albeit 60 miles north of Salim Shah in the Badakhshan town of Baharak (map here). This article gives you an idea of what things are probably like in Salim Shah.

The living conditions in this region are simply appalling. This is a region where there are no roads, no communications, no water, no power, and basically, no government, along with a population engaged in constant warring and feuding.

On Dr. Simpson’s trip to the Wakhan Corridor, his Afghan guides decided to go fishing – with explosives!

Results of fishing with explosives in the Wakhan Corridor.

50% of the children in Badakhshan are malnourished and many women die in childbirth.

A variety of terrible diseases regularly ravage large sectors of the population, and cures or even treatments are typically nonexistent. Three people were stoned to death while he was there. For Afghanistan as a whole, infant and maternal mortality figures are amongst the worst on Earth (Afghanistan’s maternal mortality rate is exceeded only by Sierra Leone) and life expectancy is a mere 45 years.

In Baharak, all of those figures are worse, if you can imagine that. The linked article includes disgusting photos of Afghans afflicted with the bizarre and terrible illnesses that ravage this region.

Moving further along the Wakhan Corridor, an area even poorer and more devastated than the rest of Badakhshan, if you can believe that, we note that on August 17, 2006 bin Laden and Zawahiri were reportedly in the Pakistani town of Darkot, on the border between Pakistan and the Wakhan Corridor of Afghanistan.

Another precarious bridge, this time near Darkot and the Wakhan Corridor. People are using yaks to cross this bridge.

The article suggested that the leadership had just moved out of Chitral to the Wakhan Corridor.

Crossing Darkot Pass on the border of Pakistan and Afghanistan. Remind you of Tora Bora and Al Qaeda’s escape through similar snowfields and peaks? Yes, it does.

Darkot Pass, another view, where the Al Qaeda leadership was reportedly hiding out in August 2006.

On a map, Darkot is north of Mir Wali and slightly northwest of the 21,407 ft. Daspar Mountain. It is located on the edge of the Wakhan Corridor on the Afghan border.

Trekkers riding yaks near Darkot. The preferred way to travel here is by yak! The Al Qaeda leadership was said to be here in summer 2006.

The entire Wakhan Corridor is at least 9000 feet high and the climate is very harsh. There is only 1 clinic in the entire corridor and it is 20 days’ donkey ride away from the furthest villages in the corridor. There is absolutely no infrastructure whatsoever here.

A very large percentage of the population is using opium here, possibly because it is about the only “medicine” available.

Dr. Simpson in an opium field in Badakhshan.

The vast majority of the population also appears to be depressed. None of the children are immunized against anything.

On Simpson’s visit, he reported that in one village, the last foreigner had passed through 20 years ago. In another, an entire village of 300 had been devastated by a klebsiella bacterial infection, probably spread by poor sanitation. At one point, there were hardly any villagers strong enough to even fetch water for the rest. The outbreak could easily have killed the whole village.

Dr. Simpson traveling along the Wakhan Corridor. The health and general living conditions here are truly horrible. This photo looks like the Andes.

A similar report, this one from a month earlier, noted that Pakistani authorities had, on July 19, 2006, ordered the entire part of Pakistan bordering the Wakhan Corridor evacuated due to reports that bin Laden and Zawahiri were in the area of the Chillinji (or Chilinji) Pass.

A raging, muddy, torrential, flooding river near Darkot. The wild, nearly unknown lands near Chillinji look similar. This entire area was evacuated in summer 2006 after reports that bin Laden and Zawahiri were hiding somewhere here. I would expect to see an Abominable Snowman in a place like this. This photo actually looks like Patagonia in Chile.

The town of Chillinji is about 34 miles west of Darkot. It is across the border of the Chitral District, beyond the tail of the Hindu Kush, into an obscure range called the Hindu Raj, and into the Ghizar District of Kashmir at the headwaters of the Ishkuman River.

A village near Darkot, one of the highest villages in the area. The terrain around Chillinji is very similar. The first modern expedition to Chillinji only explored this area in the 1990’s.

This 17,549 ft. pass is the 5th highest pass in the Hindu Kush. A report places the Al Qaeda leadership in Chillinji about 1 month before they were seen in Darkot.

The Darkot Glacier, viewed from the West. Take your breath away.

Chillinji is 30 miles east of Darkot and also borders the Wakhan Corridor.

Herding yaks near Darkot in terrain reminiscent of Tibet. Was the Al Qaeda leadership here in summer 2006?

The area has two advantages for Al Qaeda – US forces are not permitted there, and the proximity to China means that bombing by US air power is probably not an option due to the fact that a diplomatic row may ensue over US planes bombing targets so close to Chinese territory.

A precarious bridge over a raging, muddy torrent bashing away at an apparent glacier near Darkot. Tell me how a modern army travels through this terrain?

There were 100’s of tourists in the area at the time, mostly trekkers, and all were reportedly evacuated. The report was confirmed by Western intelligence in Islamabad.

The linked report above goes on to speculate about a Chinese connection to Al Qaeda, either governmental, or (more likely) via East Turkestan separatists, who are long known to be allied with Al Qaeda through attending courses at Al Qaeda’s training camps under the Taliban’s rule.

We seem to have come full circle here. Soon after the US invasion of Afghanistan, US Special Forces were hunting bin Laden in the very same Wakhan Corridor. At the time, the theory was that the top AQ leadership was hiding in caves in the corridor in a cave complex that was supposedly built by the Russians.

This complex very high in the mountains was supposedly so high-tech that it was heated. Special Forces were using thermal sensors in the winter to look for a possible heated shelter in this area. Nothing ever turned up, and no one has ever conclusively shown that this supposed base even exists.

Close to Darkot. The peaks in the background look like the Himalayas. You see Osama anywhere around here? Neither do I.

I was unable to find pictures of the Chilinji area that were not copyrighted, and we typically don’t run copyrighted photos on this blog.

Those interested in some truly spectacular photos of the Chillinji/Darkot Pass area are urged to check out the site of photographer Steve Razzetti. Once inside, click on the slide show marked Hindu Kush. Most of the photos are of the area around Chillinji and Darkot. The Chilinji area looks quite a bit like the Darkot photos in this post.

Near Darkot, the soaring peaks bring to mind the Matterhorn in the Alps or Everest in the Himalayas.

Our final stop along this voyage will be a place called Murkushi ( map), a town located 4 1/2 miles south of the Chinese border where China, Pakistan and Afghanistan all come together.

A breathtaking photo of the area near Murkushi, with a typical barely passable bridge and cascading gorge. This area is impassible by motor vehicles and often even pack animals because they are afraid of these bridges. Hence, the only access is by foot.

Murkushi is located south of Kilik Pass and west of Mintaka Pass. Both passes are at 15,449 feet.

The incredible view of Kilik Pass. Kilik Pass is only about 5 miles north of the supposed Al Qaeda training camp at Murkushi.

Mintaka Pass, about 5 miles east of Murkushi, the site of a reported Al Qaeda training camp. Photo is from the Pakistani side. The pass is about 15,500 feet.

The stunning view, looking from Kilik Pass southeast to Mintaka Pass on the Chinese-Pakistani border. This area is so close to China that it is de facto controlled by the Chinese. Furthermore, the proximity to China means that US jets cannot bomb here for fear of setting off a world crisis with the Chinese.

According to the Baldouf article, by August 2002, Al Qaeda had set up a large base here.

A great and hard to find photo of the area right around Murkushi. Al Qaeda supposedly had a large base here in Summer 2002.

Afghan spies, pretending to be Islamists, reportedly penetrated the Murkushi and Shah Salim bases in 2002. The Murkushi area, though in Pakistani territory, is reportedly under de facto Chinese control.

Afghan intelligence concluded that the Murkushi base may be being controlled by the Chinese government, a theory the Machiavellian implications of which are presently boggling my mind. I will leave it to the reader to toss that idea around!

The people residing in Darkot, Chillinji and Murkushi are generally Wakhis, an Ismaili Shia Tajik people numbering 50,000 worldwide.

Wakhi women and girls near Darkot. Note the colorful dresses of these Ismaili Shia women. Women here are not under the restrictions of purdah one finds as one goes west towards Chitral.

Wakhi is spoken in Pakistan (9,100 speakers), China (6,000 speakers), Tajikistan (7,000 speakers) and Afghanistan (9,600 speakers). In Afghanistan, it is spoken in the Wakhan Corridor, described above.

Wakhi is related to Iranian and Tajik, but fairly distantly. It is even rather diverse within its Pamir subgroup, as this lexicon shows. Wakhi is now a written language, and 60% of Pakistani speakers can read and write it. There is also a radio program in Wakhi in Pakistan. Most male Wakhis in Pakistan can also speak Urdu.

In other places, especially Afghanistan and Tajikistan (especially), Wakhi appears to be losing ground in favor of Pashto or Tajik. In Pakistan, Wakhis refer to themselves as Tajiks. Tajik nationalists claim that Wakhi is a dialect of Tajik, but a quick look at the lexicon above shows how ludicrous that claim is.

There are also some speakers of Burushaski , an obscure language isolate (not related to any known languages) here, especially in the Hunza around Murkushi. I have studied Burushaski extensively (in particular, the work of John Bengston) and I now feel it is related to Sino-Tibetan, Basque, Ket and the North Caucasian languages like Chechen in a super-family called Dene-Caucasian), although that is a highly controversial theory.

The Bengston book I read back in the 1990’s mostly dealt with Yenisien (the Ket family), some Caucasian languages and Basque, and I felt it was quite convincing.

Khowar and Shina are also spoken in the Yasin and Ishkoman Valleys near Darkot and Chillinji.

River crossing near Darkot. This party had a jeep but they had to abandon it miles back because it was no longer possible to travel by jeep. They switched to donkeys, which they are using to ford this wide, wild river. A mechanized army is useless here; an excellent place to hide.

Shina is a widely-spoken language in northern Pakistan, spoken by 330,000 speakers.

An interesting fact is that in Chillinji, Darkot and Murkushi, the purdah system, whereby women are kept out of sight in public, pretty much vanishes, and girls and women walk freely about and anyone can look at them, whereas in most of Chitral, even in the Ismaili areas, the purdah system prevails.

Wakhi girls near Darkot. Not only do they walk around freely, but you can actually look at them and even take their photos! Gasp! Note the bright outfits and the neat jewelry hanging from them. The girl on the right has very noticeable Asian features.

You may note the focus on languages in this post. I have a Master’s Degree in Linguistics. One of my areas of interest is a controversial one – language death. The decline and extinction of so many of the world’s languages is a complex subject that goes beyond the scope of this post. But those interested may wish to check out this neat paper on language diversity, decline and death in Pakistan (pdf).

It also has a nice list of the 70 languages spoken there, with number of speakers, locations, and in some cases, notes about language shift (a process of language death whereby speakers quit speaking their native language in favor of another, usually more widely spoken language).


Filed under Afghanistan, Animals, Asia, Carnivores, China, Hindi, India, Indic, Indo-Iranian, Islam, Journalism, Kashmir, Kashmiri, Language Families, Linguistics, Mammals, Pakistan, Radical Islam, Regional, Religion, Reposts From The Old Site, Sociolinguistics, South Asia, US War in Afghanistan, War, Wild

More On The Hardest Languages To Learn – Indo-European Languages

Caution: This post is very long! It runs to 184 pages on the Web. Updated November 25, 2016.

This post will deal with how hard it is for English speakers to learn other IE languages. The English section will necessarily deal with how hard it is for non-English speakers to learn English, and as such will be less scientific. Nevertheless, there are certain things about English that tend to cause problems for many, such as phrasal verbs.

We did a post on this earlier, but it looks like we only scratched the surface. There are many webpages on this topic, and one could read about the subject for a long time, but after a while, things start getting repetitive.

This post is very good. There are more in various places on the Web.

For starters, before we do our own analysis, let’s look at what some other people came up with. This post is very good. They did a survey, and the post describes the results of the survey.

According to the survey, the nine hardest languages to learn overall were Mandarin, Hungarian, Finnish, Polish, Arabic, Hindi, Icelandic, German and Swedish.

The eight hardest languages to speak (or to pronounce correctly, specifically) were French, Mandarin, Polish, Korean, Hungarian, Arabic, Basque and Hindi.

The nine hardest languages to write were Arabic, Mandarin, Polish, French, Serbo-Croatian, Japanese, Russian, Basque and English.

How does that survey line up with the facts? Surveys are just opinions of L2 learners, and carry variant validity. For starters, let’s throw Swedish off the list altogether, as it actually seems to be a pretty easy language to learn. It’s interesting that some people find it hard, but the weight of the evidence suggests that more folks find it easy than difficult.

Mandarin, Arabic, Japanese and Russian of course use different alphabets and this is why they were rated as hard to write.

Method. 42 IE languages were examined. A literature survey, combined with interviews of various L2 language learners was conducted. In addition, 100 years of surveys on the question by language instructors was reviewed. The US military’s School of Languages in Monterey’s ratings system for difficulty of learning various languages was analyzed.

Results were collated in an impressionistic manner along a majority rules line in order to form final opinions. For example, a minority said that Portuguese or Spanish were very hard to learn, but the consensus view was that they were quite easy. In this case, the minority opinion was rejected, and the consensus view was adopted. The work received a tremendous amount of criticism, often hostile to very hostile, after publication, and many changes were made to the text.

Clearly, such a project will necessarily be more impressionistic than scientific. Scientific tests of the relative difficulty of learning different languages will have to await the development of algorithms specifically designed to measure such things. And even then, surely there will be legions of “We can’t prove anything” naysayers, as this is the heyday of the “We can’t prove anything” School of Physics Envy in Linguistics.

One common criticism was, “In Linguistics, the standard view is that there is no such thing as an easy or difficult language to learn. All languages are equally difficult or easy to learn.” Unless we are talking about children learning an L1 (and even then that’s a dubious assertion), this statement was rejected as simply untrue and exemplar of the sort of soft science (“We can’t prove anything about anything”) mushiness that has overtaken Linguistics in recent years.

Sociolinguistics and Applied Linguistics have long been nearly ruined by soft science mushiness, and in recent years, soft science “We can’t prove anything” muddleheadedness has overtaken Historical Linguistics in a horrible way. Bizarrely enough, this epidemic of Physics Envy has been clouded, as one might suspect, in claims of rigorous application of the scientific method.

But hard sciences prove things all the time. Whenever a field claims that almost nothing in the field is provable, you’re heading in the realms of Politically Correct soft science Humanities brain mush.

Results. A ratings system was designed in terms of how difficult it would be for an English-language speaker to learn the language. In the case of English, English was judged according to how hard it would be for a non-English speaker to learn the language. Speaking, reading and writing were all considered.

Ratings. Languages were rated 1-5 based on difficulty for an English speaker, easiest to hardest. 1 = easiest, 2 = moderately easy to average, 3 = average to moderately difficult, 4 = very difficult, 5 = most difficult of all.

Time needed. Time needed for an English language speaker to learn the language “reasonably well”: Level 1 languages = 3 months-1 year. Level 2 languages = 6 months-1 year. Level 3 languages = 1-2 years. Level 4 languages = 2 years. Level 5 languages = 3-4 years, but some may take longer.

Conclusion. The soft science, Politically Correct mush-speak from the swamps of Sociolinguistics currently in vogue, “All languages are equally difficult or easy for any adult to learn,” was rejected. The results of this study indicate that languages to indeed differ dramatically in how difficult they are for L2 English language learners.




Ind0-Ayran languages like Kashmiri, Hindi and especially Sanskrit are quite hard, and Sanskrit is legendary for its extreme complexity.

Central Zone
Western Hindi

The Hindi script is quite opaque to Westerners, some of whom say that Chinese script is easier. You speak one way if you are talking to a man or a woman, and you also need to take into account whether you as speaker are male or female. Gender is also as prominent as in Spanish; you have to remember whether any given noun is masculine or feminine.Hindi is definitely an IE language by its rich system of gender, case and number inflection.

The most difficult aspects of Hindi are the pronunciation and the case system. In addition, Hindi is split ergative, and not only that, but it actually has a tripartite ergative system, and the ergativity is split by tense like in Persian.

The distinction between aspirated/unaspirated and alveolar/retroflex consonants is hard for many to make. There is a four-way distinction ion the t and d sounds with aspirated/unaspirated dental and aspirated/unaspirated retroflex t‘s and d‘s. The are three different r sounds – one that sounds like the English r and two retroflex r‘s that are quite hard to make or even distinguish, especially at the end of a word. Hindi also has nasalized vowels.

If you come from a language that has case, Hindi’s case system will not be overly difficult.

In addition, there is a completely separate word for each number from 1-100, which seems unnecessarily complicated.

However, Hindi has a number of cognates with English. I am not sure if they are Indic loans into English or they share a common root going back to Proto-Indo-European (PIE).

loot plunder/destroy, English loot.
season/weather, English equivalent is monsoon
storm, English equiv. typhoon
d – something tied around the waist, English equiv. cummerbund
– literally bad name, means bad reputation. These are both cognates to the English words bad and name.
bangalaahouse, English equiv. bungalow
priest, English equiv. pundit

Nevertheless, Hindi typically gets a high score in ratings of difficult languages to learn. Based on this high score across multiple surveys, we will give it a relatively high rating.

Hindi is rated 4, very hard to learn.

Punjabi is probably harder than any other Indic language in terms of phonology because it uses tones. It’s like Hindi with tones. It has either two or three tones: high or high-falling, low or low-rising and possibly a neutral or mid tone. It is very odd for an IE language to have tones.

Punjabi is rated 4.5, very difficult.

Eastern Zone

Bengali is similar to Hindi, but it lacks grammatical gender, and that fact alone is said to make it much easier to learn. Bengali speak tend to make stereotypical gender errors when speaking in Hindi. Nevertheless, it uses the Sanskrit alphabet, and that alone makes it hard to read and write.

Bengali is rated 3.5, harder than average to learn.

Northern Zone
Eastern Pahari

Nepali is a very difficult language to learn as it has a complex grammar. It has case not for nouns themselves but for clause constituents. It has tense, aspect, and voice. Nepali has an unbelievable 11 noun classes or genders, and affixes on the verb mark the gender, number and person of the subject. It even has split ergativity, strange for an IE language.

Nepali has the odd feature, like Japanese, of having verbs have completely different positive and negative forms.

~ hoina (I am ~ I am not)
chas ~ chainas (you (intimate) are ~ you are not)
bolchu ~ boldina (I speak ~ I don’t speak)

Note the extreme differences on the conjugation of the present tense of the verb to be between 1 singular and 2 familiar singular. They look nothing like each other at all.

Adjectives decline in peculiar way. There is an inflection on adjectives that means “qualified.” So can say this by either inflecting the adjective:

dublo ~ dublai (tall ~ quite tall)
hoco ~ hocai (short ~ rather short)
rāmro ~ rāmrai (nice ~ nice enough)

or by putting the invariant qualifying adverb in front of the adjective:

ali dubloquite tall
ali hocorather short
ali rāmronice enough

Nepali gets a 4.5 rating, very difficult.

Northwestern Zone


Sinhala is also difficult but it is probably easier than most other languages in the region.

Sinhala is rated 3, average difficulty.


Sanskrit is legendary for its difficulty. It has script that goes on for long sequences in which many small individual words may be buried. You have to take apart the sequences to find the small words. However, the words are further masked by tone sandhi running everything together. Once you tease the sandhi apart, you have to deal with hundreds of compound characters in the script. Once you do those two things, you are left with eight cases, nine declensions, dual number and other fun things.

Even native speakers tend to make grammatical mistakes are admit that parts of the grammar are fiendishly difficult. There are many grammatical features that are rarely or never found in any other language. Noun declension is based on the letter than the noun ends in, for instance, nouns that end in a, e or u all decline differently. There are three genders for nouns, and those all decline differently also. Each noun has eight cases and three numbers (singular, dual and plural) so there are 24 different forms for each noun. Counting the different combinations of endings and genders (all subsumed into a sort of noun class system) there are 20 different “noun classes.”

Combining the “noun classes” with the three genders, you end up with 1,440 different regular forms that nouns can take. To make matters worse, some of the cases have different forms themselves. And there are some exceptions to these rules. The I and you pronouns decline differently, but pronouns are simple compared to nouns.

For the verbs, each verb had exist in 10 different forms of tense or mood (one from Vedic Sanskrit is no longer used). There are six tenses and four moods. The six tenses are: one present tense, two future tenses and three past tenses. The moods are: imperative, dubitive (expresses uncertainty), optative (expresses hope or offers a benediction) and a form that expresses the concept if only, then… There are two different conjugations based on who is the beneficiary of the action, you or others. There are ten different classes of verbs, each of which conjugates differently. Additionally, each verb has a different form in the singular, dual and plural and in the 1st, 2nd and 3rd persons.

Once you get past all of that, you are ready to take on the really difficult parts of the language, participles, noun derivatives and agglutination, each of which is far more complicated than the above. To add insult to injury, Sanskrit has pitch accent.

Nevertheless, the language is so mathematically precise and regular that some have said it is a perfect language for computer programming. There may not be a single irregularity in the whole language.

Sanskrit is rated 5, extremely difficult.

Western Iranian
Southwestern Iranian


Persian is easier to learn than its reputation, as some say this is a difficult language to learn. In truth, it’s difficulty is only average, and it is one of the easier IE languages to learn. On the plus side, Persian has a very simple grammar and it is quite regular. It has no grammatical gender, no case, no articles, and adjectives never change form. Its noun system is as easy as that of English. The verbal system is a bit harder than English’s, but it is still much easier than that of even the Romance languages. The phonology is very simple.

On the down side, you will have to learn Arabic script. There are many lexical borrowings from Arabic which have no semantic equivalents in Persian.

English: two (native English word) ~ double (Latin borrowing)
Note the semantic transparency in the Latin borrowing.

Persian: do (native Persian word) ~ tasneyat (Arabic borrowing)
Note the utter lack of semantic correlation in the Arabic borrowing.

Some morphology was borrowed as well:

library (has an Arabic broken plural)

It is a quite easy language to learn at the entry level, but it is much harder to learn at the advanced level, say Sufi poetry, due to difficulty in untangling subtleties of meaning.

Persian gets a 3 rating as average difficulty.

Northwestern Iranian

Kurdish is about as hard to learn as Persian, but it has the added difficulty of pharyngeals, which are very hard for English speakers to make. Like Persian, it is no gender or case, and it also has a tense split ergative system.

Kurdish gets a 4 rating, very hard to learn.

Eastern Iranian

Ossetian is a strange Iranian language that has somehow developed ejectives due to proximity of Caucasian languages which had them. An IE language with ejectives? How odd.

Ossetian gets a 4 rating, very hard to learn.


Italian is said to be easy to learn, especially if you speak a Romance language or English, but learning to order a pizza and really mastering it are two different things. Foreigners usually do not learn Italian at anywhere near a native level.

For instance, Italian has three types of tenses – simple, compound, and indefinite.

There are also various moods that combine to take tense forms – four subjunctive moods, two conditional moods, two gerund moods, two infinite moods, two participle moods and one imperative mood.

There are eight tenses in the indicative mood – recent past, remote pluperfect, recent pluperfect, preterite (remote past), imperfect, present, future, future perfect. There are four tenses in the subjunctive mood – present, imperfect, preterite and pluperfect. There are two tenses in the conditional mood – present and preterite. There is only one tense in the imperative mood – present. Gerund, participle and infinite moods all take only present and perfect tenses.

Altogether, using these mood-tense combinations, any Italian verb can decline in up to 21 different ways. However, the truth is that most Italians have little understanding of many of these tenses and moods. They do not know how to use them correctly. Hence they are often only used by the most educated people. So an Italian learner does not really need to learn all of these tenses and moods.

Italian has many irregular verbs. There are 600 irregular verbs with all sorts of different irregularities. Nevertheless, it is a Romance language, and Romance has gotten rid of most of its irregularity. The Slavic languages are much more irregular than Romance.

Counterintuitively, some Italian words are masculine in the singular and feminine in the plural. There are many different ways to say the:





Few Italians even write Italian 100% correctly. However, there is no case in Italian, as in all of Romance with the exception of Romanian.

Italian is still easier to learn than French – for evidence see the research that shows Italian children learning to write Italian properly by age six, 6-7 years ahead of French children. This is because Italian orthography is quite sensible and coherent, with good sound-symbol correspondence. Nevertheless, the orthography is not as transparent as Spanish’s.

In a similar sense, Italian changes the meaning of verbs via addition of a verbal prefix:





In these cases, you create completely new verbs via the addition of the verbal prefix to the base. Without the prefix, it is a completely different verb.

Like German and French, Italian forms the auxiliary tense with two different words: avere and essere. This dual auxiliary system is more difficult than French’s and much more difficult than German’s.

Italian is somewhat harder to learn than Spanish or Portuguese but not dramatically so. Italian has more irregularities than those two and has different ways of forming plurals, including two different ways of forming plurals that can mean different things depending on the context. This is a leftover from the peculiarities of the Latin neutral gender. The rules about when plurals end in -io or -e are opaque.

In addition, Italian pronouns and verbs are more difficult than in Spanish. Grammar rules in Spanish are simpler and seem more sensible than in Italian. Italian has the pronominal adverbs ne and se. Their use is not at all intuitive, however, they can be learned with a bit of practice.

Italian pronunciation is a straightforward, but the ce and ci sounds can be problematic. The only sounds that will give you trouble are r, gl and gn.

Italian gets a 3.5 rating, average difficulty.

Often thought to be an Italian dialect, Neapolitan is actually a full language all of its own. In Italy, there is the Neapolitan language and Neapolitan Italian, which is a dialect or “accent” of Italian. Many Italians speak with a Neapolitan accent, and it is easy for non-Neapolitans to understand. However, the Neapolitan language is a a full blown language and is nearly incomprehensible to even speakers of Standard Italian.Neapolitan is said to be easier than Standard Italian. Unlike Italian, Neapolitan conjugation and the vocative are both quite simple and any irregularities that exist seem to follow definite patters.

Neapolitan gets a 2.5 rating, fairly easy.

Western Romance

French is pretty easy to learn at a simple level, but it’s not easy to get to an advanced level. For instance, the language is full of idioms, many more than your average language, and it’s often hard to figure them out.

One problem is pronunciation. There are many nasal vowels, similar to Portuguese. The eu, u and all of the nasal vowels can be Hell for the learner. There is also a strange uvular r. The dictionary does not necessarily help you, as the pronunciation stated in the dictionary is often at odds with what you will find on the street.

There are phenomena called élision, liaison and enchainement, which is similar to sandhi in which vowels elide between words in fast speech. There are actually rules for this sort of thing, but the rules are complicated, and at any rate, for liaisons at least, they are either obligatory, permitted or forbidden depending on the nature of the words being run together, and it is hard to remember which category various word combinations fall under.

The orthography is also difficult since there are many sounds that are written but no longer pronounced, as in English. Also similar to English, orthography does not line up with pronunciation. For instance, there are 13 different ways to spell the o sound: o, ot, ots, os, ocs, au, aux, aud, auds, eau, eaux, ho and ö.

In addition, spoken French and written French can be quite different. Spoken French uses words and phrases such as c’est foututhe job will not be done, and on which you might never see in written French.

The English language, having no Language Committee, at least has an excuse for the frequently irrational nature of its spelling.

The French have no excuse, since they have a committee that is set up in part to keep the language as orthographically irrational as possible. One of their passions is refusing to change the spelling of words even as pronunciation changes, which is the opposite of what occurs in any sane spelling reform. So French is, like English, frozen in time, and each one has probably gone as long as the other with no spelling reform.

Furthermore, to make matters worse, the French are almost as prickly about writing properly as they are about speaking properly, and you know how they are about foreigners mangling their language.

Despite the many problems of French orthography, there are actually some rules running under the whole mess, and it is quite a bit more sensible than English orthography, which is much more chaotic.

French has a language committee that is always inventing new native French words to keep out the flood of English loans. They have a website up with an official French dictionary showing the proper native coinages to use. Another one for computer technology only is here.

On the plus side, French has a grammar that is neither simple nor difficult; that, combined with a syntax is pretty straightforward and a Latin alphabet make it relatively easy to learn for most Westerners. In addition, the English speaker will probably find more instantly recognizable cognates in French than in any other language.

A good case can be made that French is harder to learn than English. Verbs change much more, and it has grammatical gender. There are 15 tenses in the verb, 18 if you include the pluperfect and the Conditional Perfect 2 (now used only in Literary French) and the past imperative (now rarely used). That is quite a few tenses to learn, but Spanish and Portuguese have similar situations.

A good case can be made that French is harder to learn than Italian in that French children do not learn to write French properly until age 12-13, six years after Italian children.

Its grammar is much more complicated than Spanish’s. Although the subjunctive is more difficult in Spanish than in French, French is much more irregular. Like German, there are two different ways to form the auxiliary tense to have. In addition, French uses particles like y and en that complicate the grammar quite a bit.

French is one of the toughest languages to learn in the Romance family.  In many Internet threads about the hardest language to learn, many language learners list French as their most problematic language.

This is due to the illogical nature of French spelling discussed above such that the spelling of many French words must be memorized as opposed to applying a general sound-symbol correspondence rule. In addition, French uses both acute and grave accents – `´.

French gets a 3.5 rating for more than average difficulty.


Spanish is often said to be one of the easiest languages to learn, though this is somewhat controversial. Personally, I’ve been learning it off and on since age six, and I still have problems, though Spanish speakers say my Spanish is good, but Hispanophones, unlike the French, are generous about these things.

It’s quite logical, though the verbs do decline a lot with tense and number, and there are many irregular verbs, similar to French.

Compare English declensions to Spanish declensions of the verb to read.


I read
He reads


Yo leo
Tu lees
El lee
Nosotros leemos
Vosotros leéis
Ellos leen
pudísteis haber leído
hubiéremos ó hubiésemos leído

Nevertheless, Romance grammar is much more regular than, say, Polish, as Romance has junked most of the irregularity. Spanish has the good grace to lack case, spelling is a piece of cake, and words are spoken just as they are written. However, there is a sort of case left over in the sense that one uses different pronouns when referring to the direct object (accusative) or indirect object (dative).

Spanish is probably the most regular of the Romance languages, surely more regular than French or Portuguese, and probably more regular than Italian or Romanian. Pluralization is very regular compared to say Italian. There are generally only two plurals, -s and -es, and the rules about when to use one or the other are straightforward. There is only one irregular plural:

hipérbaton -> hipérbatos

This is in reference to a literary figure and you would never use this form in day to day speech.

The trilled r in Spanish often hard for language learners to make.

There is a distinction in the verb to be with two different forms, ser and estar. Non-native speakers almost never learn the use these forms as well as a native speaker. The subjunctive is also difficult in Spanish, and L2 learners often struggle with it after decades of learning.

Spanish pronunciation is fairly straightforward, but there are some sounds that cause problems for learners: j, ll, ñ, g, and r.

One good thing about Spanish is Spanish speakers are generally grateful if you can speak any of their language at all, and are very tolerant of mistakes in L2 Spanish speakers.

Spanish is considered to be easier to learn for English speakers than many other languages, including German. This is because Spanish sentences follow English sentence structure more than German sentences do. Compared to other Romance languages, Spanish one of the easiest to learn. It is quite a bit easier than French, moderately easier than Literary Portuguese, and somewhat easier than Italian.

Nevertheless, Hispanophones say that few foreigners end up speaking like natives. Part of the reason for this is that Spanish is very idiomatic and the various forms of the subjunctive make for a wide range of nuance in expression. Even native speakers make many mistakes when using the subjunctive in conditional sentences. The dialects do differ quite a bit more than most people say they do. The dialects in Latin America and Spain are quite different, and in Latin America, the Argentine and Dominican dialects are very divergent.

Spanish gets rated 2.5, fairly easy.


Portuguese, like Spanish, is also very easy to learn, though Portuguese pronunciation is harder due to the unusual vowels such as nasal diphthongs and the strange palatal lateral ʎ, which many English speakers will mistake for an l.

Of the nasal diphthongs, ão is the hardest to make. In addition, Brazilian (Br) Portuguese has an r that sounds like an h, and l that sounds like a w and a d that sounds like a j, but only some of the time! Fortunately, in European (Eu) Portuguese, all of these sounds sound as you would expect them to.

Portuguese has two r sounds, a tapped r (ɾ) that is often misconceived as a trilled r (present in some British and Irish English dialects) and an uvular r (ʁ) which is truly difficult to make. However, this is the typical r sound found in French, German, Danish and Hebrew, so if you have a background in one of those languages, this should be an easy sound.  L2 learners not only have a hard time making them but also mix them up sometimes.

You can run many vowels together in Portuguese and still make a coherent sentence. See here:

É o a ou o b? [Euaoube]
Is it (is your answer) a or b?

That utterance turns an entire sentence into a single verb via run-on vowels, five of them in a row.

Most Portuguese speakers say that Portuguese is harder to learn than Spanish, especially the variety spoken in Portugal. Eu Portuguese elides many vowels and has more sounds per symbol than Br Portuguese does. Portuguese has both nasal and oral vowels, while Spanish has only oral values. In addition, Portuguese has 12 vowel phonemes to Spanish’s five.

Portuguese has also retained the archaic subjunctive future which has been lost in many Romance languages.

Try this sentence: When I am President, I will change the law.

In Spanish, one uses the future tense as in English:

Cuando yo soy presidente, voy a cambiar la ley.

In Portuguese, you use the subjunctive future, lost in all modern Romance languages and lacking in English:

Quando eu for presidente, vou mudar a lei. – literally, When I may be President, I will possibly change the law.

The future subjunctive causes a lot of problems for Portuguese learners and is one of the main ways that it is harder than Spanish.

There is a form called the personal infinitive in Eu Portuguese in which the infinitive is actually inflected that also causes a lot of problems for Portuguese learners.

Personal infinitive:

para eu cantar      for me to sing
para tu cantares    for you to sing
para el cantar      for him to sing
para nos cantarmos  for us to sing
para eles cantarem  for them to sing

Some sentences with the personal infinitive:

Ficamos em casa do Joao ao irmos ao Porto.
We are staying at John’s when we go to Porto.

Comprei-te um livro para o leres.
I bought you a book for you to read.

In addition, when making the present perfect in Spanish, it is fairly easy with the use have + participle as in English.

Compare I have worked.

In Spanish:

Yo he trabajado.

In Portuguese, there is no perfect to have nor is there any participle, instead, present perfect is formed via a conjugation that varies among verbs:

Eu trabalhei – because Eu hei trabalhado makes no sense in Portuguese.

Portuguese still uses the pluperfect tense quite a bit, a tense that gone out or is heading out of most IE languages. The pluperfect is used a lot less now in Br Portuguese, but it is still very widely used in Eu Portuguese. The pluperfect is used to discuss a past action that took place before another past action. An English translation might be:

He had already gone by the time she showed up.

The italicized part would be the equivalent to the pluperfect in English.

O pássaro voara quando o gato pulou sobre ele para tentar comê-lo.
The bird had (already) flown away when the cat jumped over it trying to eat it.

Even Br Portuguese has its difficulties centering around diglossia. It is written in 1700’s Eu Portuguese, but in speech, the Brazilian vernacular is used. Hence:

I love you

Amo-te or Amo-o [standard, written]
Eu te amo or Eu amo você  [spoken]

We saw them

Vimo-los [standard, written]
A gente viu eles  [spoken]

Even Eu Portuguese native speakers often make mistakes in Portuguese grammar when speaking. Young people writing today in Portuguese are said to be notorious for not writing or speaking it properly. The pronunciation is so complicated and difficult that even foreigners residing in Portugal for a decade never seem to get it quite right. In addition, Portuguese grammar is unimaginably complicated. There are probably more exceptions than there are rules, and even native speakers have issues with Portuguese grammar.

Portuguese gets a 3 rating, average difficulty.

Eastern Romance

Surprisingly enough, Romanian is said to be one of the harder Romance languages to speak or write properly. Even Romanians often get it wrong. One strange thing about Romanian is that the articles are attached to the noun as suffixes. In all the rest of Romance, articles are free words that precede the noun.

English  telephone the telephone
Romanian telefon   telefonul

Romanian is the only Romance language with case. There are five cases – nominative, accusative, genitive, dative, vocative – but vocative is not often use, and the other four cases combine as two cases: nominative/accusative and dative/genitive merge as single cases.

Nominative-Accusative aeroportul
Genitive-Dative       aeroportului

The genitive is hard for foreigners to learn as is the formation of plurals. The ending changes for no apparent reason when you pluralize a noun and there are also sound changes:

brad (singular)
brazi (plural)

Many native speakers have problems with plurals and some of the declensions. Unlike the rest of Romance which has only two genders, masculine and feminine, Romanian has three genders – masculine, feminine and neuter (the neuter is retained from Latin). However, neuter gender is realized on the surface as masculine in the singular and feminine in the plural, unlike languages such as Russian where neuter gender is an entirely different gender.

The pronunciation is not terribly difficult, but it is hard to learn at first. For some odd reason, the Latinization is considered to be terrible.

Romanian is harder to learn than Spanish or Italian and possibly harder than French. However, you can have odd sentences with nothing but vowels as in Maori.

Aia-i oaia ei, o iau eu?
That’s her sheep, should I take it?

It may have the most difficult grammar in Romance. Romanian has considerable Slavic influence and this will make it harder for the English speaker to learn than other Romance languages.

Romanian gets a 3.5 rating, more than average difficulty.

West Germanic

People often say that English is easy to learn, but that is deceptive. For one thing, English has anywhere from 500,000-1 million words (said to be twice as much as any other language – but there are claims that Dutch and Arabic each have 4 million words), and the number increases by the day. Furthermore, most people don’t understand more than 50,000, and a majority might only understand 30,000 words. Yet your average person only uses 5,000 at most.

Actually, the average American or Brit uses a mere 2,500 words. As we might expect, our cultivated Continentals in Europe, such as Spaniards and French, probably have twice the regular vocabulary of English speakers and far more colloquial expressions.

In addition, verbal phrases or phrasal verbs are a nightmare. Phrasal verbs are probably left over from “separable verbs” in German. In most of the rest of IE, these become affixes as in Latin Latin cum-, ad-, pro-, in-, ex-, etc.. In many cases, phrasal verbs can have more than 10 different antagonistic meanings.

Here is a list of 123 phrasal verbs using the preposition up after a verb:

Back up – to go in reverse, often in a vehicle, or to go back over something previously dealt with that was poorly understood in order to understand it better.
Be up – to be in a waking state after having slept. I’ve been up for three hours. Also to be ready to do something challenging. Are you up for it?
Beat up
– to defeat someone thoroughly in a violent physical fight.
Bid up – to raise the price of something, usually at an auction, by calling out higher and higher bids.
Blow up – to explode an explosive or for a social situation to become violent and volatile.
Bone up – to study hard.
Book up – all of the booking seats have been filled for some entertainment or excursion.
Bottle up – to contain feelings until they are at the point of exploding.
Break up – to break into various pieces, or to end a relationship, either personal or between entitles, also to split a large entity, like a large company or a state.
Bruise up – to receive multiple bruises, often serious ones.
Brush up – to go over a previously learned skill.
Build up – to build intensively in an area, such as a town or city, from a previously less well-developed state.
Burn up – burn completely or to be made very angry.
Bust up – to burst out in laughter.
Buy up – to buy all or most all of something.
Call up – to telephone someone. Or to be ordered to appear in the military. The army called up all males aged 18-21 and ordered them to show up at the nearest recruiting office.
Catch up
– to reach a person or group that one had lagged behind earlier, or to take care of things, often hobbies, that had been put off by lack of time.
Chat up – to talk casually with a goal in mind, usually seduction or at least flirtation.
Cheer up – to change from a downcast mood to a more positive one.
Chop up – to cut into many, often small, pieces.
Clam up – to become very quiet suddenly and not say a thing.
Clean up – to make an area thoroughly tidy or to win completely and thoroughly.
Clear up – for a storm to dissipate, for a rash to go away, for a confusing matter to become understandable.
Close up – to close, also to end business hours for a public business.
Come up – to approach closely, to occur suddenly or to overflow.
Cook up – to prepare a meal or to configure a plan, often of a sly, ingenious or devious nature. They cooked up a scheme to swindle the boss.
Crack up
– to laugh, often heartily or to fall apart emotionally.
Crank up – elevate the volume.
Crawl up – to crawl inside something.
Curl up – to rest in a curled body position, either alone or with another being.
Cut up – to shred or to make jokes, often of a slapstick variety.
Do up – apply makeup to someone, often elaborately.
Dream up – to imagine a creative notion, often an elaborate one.
Dress up – to dress oneself in formal attire.
Drive up – to drive towards something and then stop, or to raise the price of something by buying it intensively.
Drum up – to charge someone with wrongdoing, usually criminal, usually by a state actor, usually for false reasons.
Dry up – to dessicate.
Eat up – implies eating something ravenously or finishing the entire meal without leaving anything left.
End up – to arrive at some destination after a long winding, often convoluted journey either in space or in time.
Face up – to quit avoiding your problems and meet them head on.
Feel up – to grope someone sexually.
Get up – to awaken or rise from a prone position.
Give up – to surrender, in war or a contest, or to stop doing something trying or unpleasant that is yielding poor results, or to die, as in give up the ghost.
Grow up – to attain an age or maturity or to act like a mature person, often imperative.
Hang up – to place on a hanger or a wall, to end a phone call.
Hike up – to pull your clothes up when they are drifting down on your body.
Hit up – to visit someone casually or to ask for a favor or gift, usually small amounts of money.
Hold up – to delay, to ask someone ahead of you to wait, often imperative. Also a robbery, usually with a gun and a masked robber.
Hook up – to have a casual sexual encounter or to meet casually for a social encounter, often in a public place; also to connect together a mechanical devise or plug something in.
Hurry up – imperative, usually an order to quit delaying and join the general group or another person in some activity, often when they are leaving to go to another place.
Keep up – to maintain on a par with the competition without falling behind.
Kiss up – to mend a relationship after a fight.
Knock up – to impregnate.
Lay up – to be sidelined due to illness or injury for a time.
Let up – to ease off of someone or something, for a storm to dissipate, to stop attacking someone or s.t.
Lick up – to consume all of a liquid.
Light up – to set s.t. on fire or to smile suddenly and broadly.
Lighten up – to reduce the downcast or hostile seriousness of the mood of a person or setting.
Listen up – imperative – to order someone to pay attention, often with threats of aggression if they don’t comply.
Live up – to enjoy life.
Lock up – to lock securely, often locking various locks, or to imprison, or for an object or computer program to be frozen or jammed and unable to function.
Look up – to search for an item of information in some sort of a database, such as a phone book or dictionary. Also to admire someone.
Make up – to make amends, to apply cosmetics to one’s face or to invent a story.
Man up – to elevate oneself to manly behaviors when one is slacking and behaving in an unmanly fashion.
Mark up – to raise the price of s.t.
Measure up – in a competition, for an entity to match the competition.
Meet up – to meet someone or a group for a get meeting or date of some sort.
Mess up – to fail or to confuse and disarrange s.t. so much that it is bad need or reparation.
Mix up – to confuse, or to disarrange contents in a scattered fashion so that it does not resemble the original.
Mop up – mop a floor or finish off the remains of an enemy army or finalize a military operation.
Move up – to elevate the status of a person or entity in competition with other entities- to move up in the world.
Open up – when a person has been silent about something for a long time, as if holding a secret, finally reveals the secret and begins talking.
Own up – to confess to one’s sins under pressure and reluctantly.
Pass up – to miss an opportunity, often a good one.
Patch up – to put together a broken thing or relationship.
Pay up – to pay, usually a debt, often imperative to demand payment of a debt, to pay all of what one owes so you don’t owe anymore.
Pick up – to grasp an object and lift it higher, to seduce someone sexually or to acquire a new skill, usually rapidly.
Play up – to dramatize.
Pop up – for s.t. to appear suddenly, often out of nowhere.
Put up – to hang, to tolerate, often grudgingly, or to put forward a new image.
Read up – to read intensively as in studying.
Rev up – to turn the RPM’s higher on a stationary engine.
Ring up – to telephone someone or to charge someone on a cash register.
Rise up – for an oppressed group to arouse and fight back against their oppressors.
Roll up – to roll s.t. into a ball, to drive up to someone in a vehicle or to arrest all the members of an illegal group. The police rolled up that Mafia cell quickly.
Run up
– to tally a big bill, often foolishly or approach s.t. quickly.
Shake up – to upset a paradigm, to upset emotionally.
Shape up – usually imperative command ordering someone who is disorganized or slovenly to live life in a more orderly and proper fashion.
Shoot up – to inject, usually illegal drugs, or to fire many projectiles into a place with a gun.
Show up – to appear somewhere, often unexpectedly.
Shut up – to silence, often imperative, fighting words.
Sit up – to sit upright.
Slip up – to fail.
Speak up – to begin speaking after listening for a while, often imperative, a request for a silent person to say what they wish to say.
Spit up – to vomit, usually describing a child vomiting up its food.
Stand up – to go from a sitting position to a standing one quickly.
Start up – to initialize an engine or a program, to open a new business to go back to something that had been terminated previously, often a fight; a recrudescence.
Stay up – to not go to bed.
Stick up – to rob someone, usually a street robbery with a weapon, generally a gun.
Stir up – stir rapidly, upset a calm surrounding or scene or upset a paradigm.
Stop up – to block the flow of liquids with some object(s).
Straighten up – to go from living a dissolute or criminal life to a clean, law abiding one.
Suck up – to ingratiate oneself, often in an obsequious fashion.
Suit up – to get dressed in a uniform, often for athletics.
Sweep up – to arrest all the members of an illegal group, often a criminal gang.
Take up – to cohabit with someone – She has taken up with him. Or to develop a new skill, to bring something to a higher elevation, to cook something at a high heat to where it is assimilated.
Talk up – to try to convince someone of something by discussing it dramatically and intensively.
Tear up – to shred.
Think up – to conjure up a plan, often an elaborate or creative one.
Throw up – to vomit.
Touch up – to apply the final aspects of a work nearly finished.
Trip up – to stumble mentally over s.t. confusing.
Turn up – to increase volume or to appear suddenly somewhere.
Vacuum up – to vacuum.
Use up – to finish s.t. completely so there is no more left.
Wait up – to ask other parties to wait for someone who is coming in a hurry.
Wake up – to awaken.
Walk up – to approach someone or something.
Wash up – to wash.
Whip up – to cook a meal quickly or for winds to blow wildly.
Work up – to exercise heavily, until you sweat to work up a sweat. Or to generate s.t. a report or s.t. of that nature done rather hurriedly in a seat of the pants and unplanned fashion. We quickly worked up a formula for dealing with the matter.
Wrap up
– To finish something up, often something that is taking too long. Come on, let us wrap this up and getting it over with. Also, to bring to a conclusion that ties the ends together. The story wraps up with a scene where they all get together and sing a song.
Write up
– often to write a report of reprimand or a violation. The officer wrote him for having no tail lights.

Here are  phrasal verbs using the preposition down:

Back down – to retreat from a challenge or a threat.
Be down  – to be ready to ready to do something daring, often s.t. bad, illegal or dangerous, such as a fight or a crime. Are you down?
Blow down – to knock something down via a strong wind.
Break down – to take anything apart in order to reveal its component parts.
Burn down
– reduce s.t. to ashes, like a structure.
Chop down – to fell a tree with an ax.
Clamp down – to harshly police something bad in order to reduce its incidence, especially s.t. that had been ignored in the past.
Climb down – to retract a poorly made statement.
Cook down – to reduce the liquid content in a cooked item.
Crack down – To police harshly against people doing bad things.
Cut down – to fell a tree by any means or to reduce the incidence of anything, especially something bad.
Drink down – to consume all of s.t.
Drive down – to harshly bring down the price of something, often through brutal means. Investors drove down the price of the stock after the company’s latest product failed badly.
Dress down – to deliberately dress more poorly than expected, often as a trendy fashion statement.
Get down – to have fun and party, or to lie prone and remain there or to reduce something to bare essentials. Get down on the floor or Getting down to brass tacks, how can we possibly explain this anomaly other than in this particular manner?
Hang down – to let one’s hair fall down in front of one’s eyes or to hang s.t. often a banner, from a building or structure.
Hike down – to lower one’s pants. The gangsters hike their pants down to look tough.
Hold down – to hold someone or s.t. on the floor so they cannot rise or get up.
Keep down – to prevent a group, often a repressed group, from achieving via oppression by a ruling group. The Whites are keeping us Black people down.
Kick down – Drug slang meaning to contribute your drugs to a group drug stash so others can consume them with you, to share your drugs with others. Often used in a challenging sense.
Knock down – to hit or strike something so hard that it falls to the ground or collapses.
Let down – to be discouraged by something one had high hopes for.
Live down – to recover from a humiliating experience. After he was publicly humiliated, he was never able to live down his rejection by the people.
Look down – to regard someone in a negative or condemnatory way from a the point of a superior person.
Mark down – to discount the price of s.t., often significantly.
Party down – to have fun and party
Pass down – to leave s.t. of value to someone as an inheritance after a death or to inherit a saying or custom via one’s ancestors through time. It was passed down through the generations.
Pat down – to frisk.
Pay down – to reduce a bill, often a large bill, by making payments, often significant payments. We are slowly paying down that bill.
Play down – to reduce the significance of s.t. often s.t. negative, often in order to deceive people into thinking s.t. is better than it really is.
Put down – to criticize someone in a condescending way as a superior person, to insult.
Play down – to deemphasize.
Rip down – to tear s.t. off of a wall such as a sheet or poster.
Run down – to run over something or someone with a vehicle, to review a list or to attack someone verbally for a long time.
Set down – to postulate a set of rules for something.
Shake down – to rob someone purely through the use of verbal or nonphysical force or power.
Shoot down – to shoot at a flying object like a plane, hitting it so it crashes to the ground or to reject harshly a proposal.
Shut down – to close operations of an entity.
Speak down – to talk to someone in a condescending way from the point of view of a superior person.
Take down – to demolish s.t. like a building, to tackle someone, or to raid and arrest many members of an illegal organization.
Talk down – to speak to someone in an insulting manner as if one was superior or to mollify a very angry person to keep them from causing future damage. The police were able to talk down the shooter until he laid down his fun and set the hostages free.
Tear down – to demolish or destroy someone verbally or to destroy s.t. by mechanical means.
Throw down – to throw money or tokens into the pile in the center when gambling.
Turn down – to reduce the volume of something or to reject an offer.
Write down – to write on a sheet of paper

There are figures of speech and idioms everywhere (some estimate that up to 20% of casual English speech is idiomatic), and it seems impossible to learn them all. In fact, few second language learners get all the idioms down pat.

The spelling is insane and hardly follows any rules at all. The English spelling system in some ways is frozen at about the year 1500 or so. The pronunciation has changed but the spelling has not. Careful studies have shown that English-speaking children take longer to read than children speaking other languages (Finnish, Greek and various Romance and other Germanic languages) due to the difficulty of the spelling system. Romance languages were easier to read than Germanic ones.

This may be why English speakers are more likely to be diagnosed dyslexic than speakers of other languages. The dyslexia still exists if you speak a language with good sound-symbol correspondence, but it’s covered up so much by the ease of the orthography that it seems invisible, and the person can often function well. But for a dyslexic, trying to read English is like walking into a minefield.

Letters can make many different sounds, a consequence of the insane spelling system. A single sound can be spelled in many different ways: e can be spelled e, ea, ee, ei, eo, ey, ae, i, ie, and y. The k sound can spelled as c, cc, ch, ck, k, x, and q.

The rules governing the use of the indefinite, definite and zero article are opaque and possibly don’t even exist. There are synonyms for almost every word in a sentence, and the various shades of meaning can be difficult to discern. In addition, quite a few words have many different meanings. There are strange situations like read and read, which are pronounced differently and mean two different things.

English word derivation is difficult to get your mind around because of the dual origins of the English language in both Latin/French and German.

See and hear and perceptible and audible mean the same thing, but the first pair is derived from German, and the second pair is derived from Latin.

English word derivation is irregular due for the same reason:

assumeassumption (Latin)
childchildish (German)
buildbuilding (German)

In English we have at least 12 roots with the idea of two in them:


However, English regular verbs generally have only a few forms in their normal paradigm. In this arrangement, there are only five forms of the verb in general use for the overwhelming majority of verbs:

present except 3rd singular  steal
3rd person singular          steals
progressive                  stealing
past                         stole
perfect                      stolen

Even a language like Spanish has many more basic forms than that. However, coming from an inflected language, the marking of only the 3rd singular and not marking anything else may seem odd.

The complicated part of English verbs is not their inflection – minimal as it is – but instead lies in the large number of irregular verbs.

There is also the oddity of the 2nd person being the same in both the singular and the plural – you. Some dialects such as US Southern English do mark the plural – you all or y’all.

English prepositions are notoriously hard, and few second language learners get them down right because they seem to obey no discernible rules.

One problem that English learners complain of is differential uses of have.

  1. Perfect tense. I have done it.
  2. Deontic (must). I have to do it.
  3. Causative. I had it done.

While English seems simple at first – past tense is easy, there is little or no case, no grammatical gender, little mood, etc., that can be quite deceptive. In European countries like Croatia, it’s hard to find a person who speaks English with even close to native speaker competence.

There are quite a few English dialects – over 100 have been recorded in London alone.

The problem with English is that it’s a mess! There are languages with very easy grammatical rules like Indonesian and languages with very hard grammatical rules like Arabic. English is one of those languages that is simply chaotic. There are rules, but there are exceptions everywhere and exceptions to the exceptions. Grammatically, it’s disaster area. It’s hard to know where to start.

However, it is often said that English has no grammatical rules. Even native speakers make this comment because that is how English seems due to its highly irregular nature. Most English native speakers, even highly educated ones, can’t name one English grammatical rule. Just to show you that English does have rules though, I will list some of them.

*Indicates an ungrammatical form.

Adjectives appear before the noun in noun phrases:

Small dogs barked.
*Dogs small barked.

Adjectives are numerically invariant:

the small dog
the small dogs
The dog is small.
The dogs are small.

Intensifiers appear before both attributive and predicative adjectives:

The very small dog barked.
*The small very dog barked.

The dog was very small.
*The dog was small very.

Attributive adjectives can have complements:

The dog was scared.
The dog was scared of cats.

But predicative adjectives cannot:

The scared dog barked.
*The scared of cats dog barked.

Articles, quantifiers, etc. appear before the adjective (and any intensifier) in a noun phrase:

The very small dog barked.
*Very the small dog barked.
*Very small the dog barked.

Every very small dog barked.
*Very every small dog barked.
*Very small every dog barked.

Relative clauses appear after the noun in a noun phrase:

The dog that barked.
*The that barked dog.

The progressive verb form is the bare form with the suffix -ing, even for the most irregular verbs in the language:



The infinitive verb form is to followed by the bare form, even for the most irregular verbs in the language:

to be
to have
to do

*to was
*to are
*to am.

The imperative verb form is the bare form, even for the most irregular verb in the language:



All 1st person present, 2nd person present, and plural present verb forms are equivalent to the bare form, except for to be.

All past tense verb forms of a given verb are the same regardless of person and number, except for to be.

Question inversion is optional:

You are leaving?
Are you leaving?

But when inversion does occur in a wh-question, a wh-phrase is required to be fronted:

You’re seeing what?
What are you seeing?

*Are you seeing what?

Wh-fronting is required to affect an entire noun phrase, not just the wh-word:

You are going to which Italian restaurant?
Which Italian restaurant are you going to?

*Which are you going to Italian restaurant?
*Which Italian are you going to restaurant?
*Which restaurant are you going to Italian?

Wh-fronting only happens once, never more:

What are you buying from which store
Which store are you buying what from?

*What which store are you buying from?
*Which store what are you buying from?

The choice of auxiliary verb in compound past sentences does not depend on the choice of main verb:

I have eaten.
I have arrived.

*I am eaten.
*I am arrived.

cf. French

J’ai mangé.
Je suis arrivé.

English can be seen as an inverted pyramid in terms of ease of learning. The basics are easy, but it gets a lot more difficult as you progress in your learning.

While it is relatively easy to speak it well enough to be more or less understandable most of the time, speaking it correctly is often not possible for a foreigner even after 20 years of regular use.

English only gets a 2.5 rating , somewhat difficult.

High German

German’s status is controversial. It’s long been considered hard to learn, but many learn it fairly easily.

Pronunciation is straightforward, but there are some problems with the müde, the Ach, and the two ch sounds in Geschichte. Although the first one is really an sch instead of a ch, English speakers lack an sch, so they will just see that as a ch. Further, there are specific rules about when to use the ss (or sz as Germans say) or hard s. The r in German is a quite strange ʁ, and of common languages, only French has a similar r. The çχ and ‘ü sounds can be hard to make. Consonant clusters like Herkunftswörterbuch or Herbstpflanze can be be difficult. German permits the hard to pronounce shp and shtr consonant clusters. Of the vowels, ö and ü seem to cause the most problems.

German grammar is quite complex. It recently scored as one of the weirdest languages in Europe on a study, and it also makes it onto worst grammars lists. The main problem is that everything is irregular. Nouns, plurals, determiners, adjectives, superlatives, verbs, participles – they are all irregular. It seems that everything in the language is irregular.

There are six different forms of the depending on the noun case:


but 16 different slots to put the six forms in, and the gender system is irrational. In a more basic sense and similar to Danish, there are three basic forms of the:


Each one goes with a particular noun, and it’s not very clear what the rules are.

One problem with German syntax is that the verb, verbs or parts of verbs doesn’t occur until the end of the sentence. This sentence structure is known as V2 syntax, and it is quite alien for English speakers. There are verbal prefixes, and they can be modified in all sorts of ways that change meanings in a subtle manner. There are dozens of different declension types for verbs, similar to Russian and Irish. There are also quite a few irregular verbs that do not fit into any of the paradigms.

German also has Schachtelsätze, box clauses, which are like clauses piled into other clauses. In addition, subclauses use SOV word order. Whereas in Romance languages you can often throw words together into a sentence and still be understood if not grammatical, in German, you must learn the sentence structure – it is mandatory and there is no way around it. The syntax is very rigid but at least very regular.

German case is also quite regular. The case exceptions can be almost counted on one hand. However, look at the verb:


in which the direct object is in dative rather than the expected absolutive.

An example of German case (and case in general) is here:

The leader of the group gives the boy a dog.

In German, the sentence is case marked with the four different German cases:

Der Führer (nominative)
der Gruppe
gibt dem Jungen (dative)
einen Hund (accusative).

There are three genders, masculine, feminine and neutral. Yet it is difficult to tell which gender any particular noun is based on looking at it, for instance, petticoat is masculine! Any given noun inflects via the four cases and the three genders. Furthermore, the genders change between masculine and feminine in the same noun for no logical reason. Gender seems to be one of the main problems that German learners have with the language. Figuring out which word gets which gender must simply be memorized as there are no good clues.

Phonology also changes strangely as the number of the noun changes:

Haushouse (singular)
Haeuserhouses (plural with umlaut)

But to change the noun to a diminutive, you add -chen:

Haueschen – little house (singular, yet has the umlaut of the plural)

This is part of a general pattern in Germanic languages of roots changing the vowel as verbs, adjectives and nouns with common roots change from one into the other. For instance, in English we have the following vowel changes in these transformed roots:

foul filth
tell tale
long length
full fill
hot  heat
do   does

Much of this has gone out of English, but it is still very common in German. Dutch is in between English and German.


For sick, we have:

krank      sick
kränker    sicker
kränklich  sickly
krankhaft  pathological
kranken an to suffer from
kränken    to hurt
kränkeln   to be ailing
erkranken  to fall ill

For good, we have:

gut     good
Güte    goodness
Gut     a good
Güter   goods
gütig   kind
gütlich amicable

German also has a complicated preposition system.

German also has a vast vocabulary, the fourth largest in the world. This is either positive or negative depending on your viewpoint. Language learners often complain about learning languages with huge vocabularies, but as a native English speaker, I’m happy to speak a language with a million words. There’s a word for just about everything you want to say about anything, and then some!

On the plus side, word formation is quite regular.

Pollution is Umweltverschmutzung. It consists, logically, of two words, Umwelt and Verschmutzung, which mean environment and dirtying.

In English, you have three words, environment, dirtying and pollution, the third one, the combination of the first two, has no relation to its semantic roots in the first two words.

Nevertheless, this has its problems, since it’s not simple to figure out how the words are stuck together into bigger words, and meanings of morphemes can take years to figure out.

German has phrasal verbs as in English, but the meaning is often somewhat clear if you take the morphemes apart and look at their literal meanings. For instance:

vorschlagento suggest parses out to er schlägt vorto hit forth

whereas in English you have phrasal verbs like to get over with which even when separated out, don’t make sense literally.

German, like French and Italian, has two auxiliary tenses – habe and bin. However, their use is quite predictable and the tenses are not inflected so the dual auxiliary is easier in German than in French or especially Italian.

Reading German is actually much easier than speaking it, since to speak it correctly, you need to memorize not only genders but also adjectives and articles.

German is not very inflected, and the inflection that it does take is more regular than many other languages. Furthermore, German orthography is phonetic, and there are no silent letters.

German, like Dutch, is being flooded with English loans. While this helpful to the English speaker, others worry that the language is at risk of turning into English.

Learning German can be seen as a pyramid. It is very difficult to grasp the basics, but once you do that, it gets increasingly easy as the language follows relatively simple rules and many words are created from other words via compound words, prefixes and suffixes.

Rating German is hard to do. It doesn’t seem to deserve to a very high rating, but it makes a lot of people’s “hardest language you ever tried to learn” list for various reasons.

German gets a 3.5 rating, moderately difficult.

Low Franconian

While Dutch syntax is no more difficult than English syntax, Dutch is still harder to learn than English due to the large number of rules used in both speaking and writing. The Dutch say that few foreigners learn to speak Dutch well. Part of the problem is that some words have no meaning at all in isolation (meaning is only derived via a phrase or sentence). Word order is somewhat difficult because it is quite rigid. In particular, there are complex and very strange rules about the order of verbs in verbal clusters. It helps if you know German as the rule order is similar, but Dutch word order is harder than German word order. Foreigners often seem to get the relatively lax Dutch rules about word order wrong in long sentences.

Verbs can be difficult. For instance, there are no verbs get and move. Instead, get and move each have about a dozen different verbs in Dutch. A regular Dutch verb has six different forms.

Dutch spelling is difficult, and most Dutch people cannot even spell Dutch correctly. There are only two genders – common and neuter – as opposed to three in German – feminine, neuter and masculine. In Dutch, the masculine and feminine merged in the common gender. But most Dutch speakers cannot tell you the gender of any individual word, in part because there are few if any clues to the gender of any given noun.

There are remnants of the three gender system in that the Dutch still use masculine/feminine for some nouns. In the Netherlands now, most Dutch speakers are simply using masculine (common) for most nouns other than things that are obviously feminine like the words mother and sister.

However, in Belgium, where people speak Flemish, not Dutch, most people still know the genders of words. Not only that but the 3-gender system with masculine, feminine and neuter remains in place in Flemish. In addition, in Flemish, the definite article still makes an obvious distinction between masculine and feminine, so it is easy to figure out the gender of a noun:

ne man, nen boom, nen ezel, nen banaan (masculine)
een vrouw, een koe, een wolk, een peer (feminine)

In addition, most Dutch speakers cannot tell you what pronoun to use in the 3rd person singular when conjugating a verb.

This is because there are two different systems in use for conjugating the 3sing.

The basic paradigm is:

hij      he
zij (ze) she
het      it

System 1
male persons    hij
female persons  zij
neuter words    het
animals         hij, unless noun = neuter
objects         hij, "       "
abstractions    zij, "       "
substances      hij, "       "

System 2
male persons      hij
female persons    zij
all animals       hij
all objects       hij
all abstractions  zij
all substances    het

For instance, melk is a common noun. Under system 1, it would be hij. But under system 2, it would be het because it is a substance.

The er word is tricky in Dutch. Sometimes it is translated as English there, but more often then not it is simply not translated in English translations because there is no good translation for it. There are two definite articles, de and het, and they are easily confused.

Dutch has something called modal particles, the meanings of which are quite obscure.

Some say Dutch is irregular, but the truth is that more than Dutch has a multitude of very complex rules, rules that are so complicated that is hard to even figure them out, much less understand them. Nevertheless, Dutch has 200 irregular verbs.

In some respects, Dutch is a more difficult language than English. For instance, in English, one can simply say:

The tree is in the garden.

But in Dutch (and also in German) you can’t say that. You have to be more specific. What is the tree doing in the garden? Is it standing there? Is it lying on the grass? You have to say not only that the tree is in the garden, but what it is doing there.

In Dutch, you need to say:

Daar ligt een boom in de tuin.
The tree is standing in the garden.

Daar ligt een boom in de tuin.
The tree is lying in the garden.

Dutch pronunciation is pretty easy, but the ui, euij, au, ou, eeuw and uu sounds can be hard to make. Dutch speakers say only Germans learn to pronounce the ui correctly.

Dutch was listed as one of the top weirdest languages in Europe in a recent study.

Dutch is almost being buried in a flood of English loans. While this helps the English speaker, others worry that the Dutch nature of the language is at risk.

Dutch seems to be easier to learn than German. Dutch has fewer cases, thus fewer articles and and adjective endings. There are two main ways of pluralizing in Dutch: adding -‘s and adding -en. Unfortunately, in German, things are much more complex than that. Dutch has only two genders (and maybe just a trace of a third) but German definitely has three genders. Verb conjugation is quite similar in both languages, but it is a bit easier in Dutch. Word order is the same: complex in both languages. Both languages are equally complex in terms of pronunciation. Both have the difficult ø and y vowels.

Dutch gets a 3 rating, average difficulty.

Afrikaans is just Dutch simplified.

Where Dutch has 200 irregular verbs, Afrikaans has only six. A Dutch verb has six different forms, but Afrikaans has only two. Afrikaans has two fewer tense than Dutch. Dutch has two genders, and Afrikaans has only one. Surely Afrikaans ought to be easier to learn than Dutch.

Afrikaans gets a 2 rating, very easy to learn.

North Germanic
West Scandinavian

Icelandic is very hard to learn, much harder than Norwegian, German or Swedish. Part of the problem is pronunciation. The grammar is harder than German grammar, and there are almost no Latin-based words in it. The vocabulary is quite archaic. Modern loans are typically translated into Icelandic equivalents rather than borrowed fully into Icelandic.

There are four cases: nominative, accusative, dative and genitive – as in German – and there are many exceptions to the case rules, or “quirky case,” as it is called. In quirky case, case can be marked on verbs, prepositions and and adjectives. The noun morphology system is highly irregular. Articles can be postfixed and inflected and added to the noun. In fact, Icelandic in general is highly irregular, not just the nouns.

Verbs are modified for tense, mood, person and number, as in many other IE languages (this is almost gone from English). There are up to ten tenses, but most of these are formed with auxiliaries as in English. Icelandic also modifies verbs for voice – active, passive and medial. Furthermore, there are four different kinds of verbs – strong, weak, reduplicating and irregular, with several conjugation categories in each division.  Many verbs just have to be memorized.

Adjectives decline in an astounding 130 different ways, but many of these forms are the same.

The language is generally SVO, but since there is so much case-marking, in poetry all possibilities – SVO, SOV, VSO, VOS, OSV and OVS – are allowed. There is also something odd called “long distance reflexives,” which I do not understand.

In addition, Icelandic has the typical Scandinavian problem of a nutty orthography.

Icelandic verbs are very regular but the sounds change so much, especially the vowels, that the whole situation gets confusing pretty fast. In addition, there are three different verbal paradigms depending on the ending of the verb:


Icelandic verbs are commonly cited as some of the hardest verb systems around, at least in Europe. Even Icelandic people say their own verbs are difficult.

Icelandic has a voiceless lateral l. This can be a hard sound to make for many learners, especially in the middle of a word. In addition, there are two alveolar trills (the rolled r sound in Spanish), and one of them is voiced while the other is voiceless. Learners say they have problems with both of these sounds. In addition to voiceless l‘s and r‘s, Icelandic also has four voiceless nasals – , , ɲ̊, and ŋ̊ – the n, m, ny (as in Spanish nina), and ng sounds.

There are also contrasts between aspirated and nonaspirated stops including the odd palatal stops and c. In addition, there is a strange voiceless palatal fricative ç (similar to the h in English huge). In addition, Icelandic has a hard to pronounce four consonant cluster strj- that occurs at the beginning of a word.

Icelandic does have the advantage of being one of the few major languages with no significant dialects, so this is a plus. Icelandic has been separated from the rest of Scandinavian for 1,100 years. Icelandic is spoken over a significant region, much of which has inhabited places separated by large expanses of uninhabitable land such as impassable glaciers, volcanoes, lava flows,  geysers and almost no food. How Icelandic managed to not develop dialects in this situation is mysterious.

Icelandic has traditionally been considered to be one of the hardest languages on Earth to learn.

Icelandic gets a 5 rating, extremely difficult to learn.

Faroese is said to be even harder to learn than Icelandic, with some very strange vowels not found in other North Germanic languages.

Faroese has strong, weak and irregular verbs. It also has a strange supine tense.

The Faroese orthography is as irrational as Icelandic’s. There are so many rules to learn to be able to write Faroese properly. Faroese, like Icelandic, prefers to coin new words rather than borrow words wholesale into its language. Therefore the English speaker will not see a lot of obvious borrowings to help them out. Some argue against this nativization process, but maybe it is better than being buried in English loans like German and Dutch are at the moment.

computertelda (derived from at telja – to count. Icelandic has a similar term.
helicoptertyrla (derived from tyril – a spinning tool for making wool or loom.
pocket calculator
telduhvølpur (Lit. computer puppy), roknimaskina (Lit. calculating machine)

Faroese has the advantage of having no verbal aspect, and verbal declension does not differ much according to person. However, Faroese has a case system like Icelandic.

Faroese gets a 5 rating,extremely difficult.

Norwegian is fairly easy to learn, and Norwegian is sometimes touted as the easiest language on Earth to learn for an English speaker.

This is confusing because Danish is described below as a more difficult language to learn, and critics say that Danish and Norwegian are the same, so they should have equal difficulty. But only one Norwegian writing system is almost the same as Danish the Danish writing system. Danish pronunciation is quite a bit different from Norwegian, and this is where the problems come in.

Even Norwegian dialects can be a problem. Foreigners get off the plane having learned a bit of Norwegian and are immediately struck by the strangeness of the multiplicity of dialects, which for the most part are easy for Norwegians to understand but can be hard for foreigners. Norwegians often only understand their many dialects due to bilingual learning and much exposure and there are definitely Norwegian dialects that even Norwegians have a hard time understand like Upper and Lower Sogn and Trondnersk.

There is also the problematic en and et alternation, as discussed with Danish. Norwegian has an irrational orthographic system, like Swedish, with silent letters and many insensible sounds, both consonants and vowels. It has gone a long time without a spelling reform. It has the additional orthographic issues of two different writing systems and a multitude of dialects. Norwegian, like Danish and Swedish, has a huge vowel inventory, one of the larger ones on Earth. It can be confusing and difficult to make all of those odd vowel sounds: 18 contrasting simple vowels, nine long and nine short , , ɛː, ɑː, , , ʉ̟ː, , øː, ɪ, ɛ, a, ɔ, ʊ, ɵ, ʏ and œ.

Norwegian has very little inflection in its words, but the syntax is very difficult. Norwegian also has “tonemes” which distinguish between homophones.

tankenthe tank
the thought

have two different meanings, even though the stress and pronunciation are the same. The words are distinguished by a toneme.

For some reason, Norwegian scored very high on a study of weirdest languages on Earth, but Swedish and Danish also got high scores.

However, Norwegian is a very regular language.

Norwegian gets a 2 rating, moderately easy to learn.

East Scandinavian

Danish is a harder language to learn than one might think. It’s not hard to read or even write, but it’s quite hard to speak. However, like English, Danish has a non-phonetic orthography, so this can be problematic. It has gone a long time without a spelling reform, so there are many silent letters and sounds, both vowels and consonants, that make no sense. Danish makes it on lists of most irrational orthographies of all.

In addition, there are d words where the d is silent and other d words where it is pronounced, and though the rules are straightforward, it’s often hard for foreigners to get the hang of this. The d in hund is silent, for instance. In addition, the b, d, and g sounds are somehow voiceless in many environments. There are also the strange labiodental glide and alveopalatal fricative sounds. In certain environments, d, g, v, and r turn into vowels.

There are three strange vowels that are not in English, represented by the letters æ, ø and å. They are all present in other Scandinavian languages – æ is present in Icelandic and Norwegian, ø is part of Norwegian, and å is part of Norwegian and Swedish, but English speakers will have problems with them. In addition, Danish has creaky-voiced vowels, which is very strange for an IE language. Danish language learners often report having a hard time pronouncing Danish vowels or even telling one apart from the other. Danish makes it onto lists of the wildest phonologies on Earth,and it made it high on a list of weirdest languages on Earth.

One advantage of all of the Scandinavian languages is that their basic vocabulary (the vocabulary needed to converse at a basic level and be understood) is fairly limited. In other words, without learning a huge number of words, it is possible to have a basic conversation in these languages. This is in contrast to Chinese, where you have to learn a lot of vocabulary just to converse at a basic level.

As with Maltese and Gaelic, there is little correlation between how a Danish word is written and how it is pronounced.

Pronunciation of Danish is difficult. Speech is very fast and comes out in a continuous stream that elides entire words. Vowels in the middle and at the end of words are seldom expressed. There are nine vowel characters, and each one can be pronounced in five or six different ways. There is nearly a full diphthong set, and somehow pharyngealization is used as an accent. Danish has a huge set of vowels, one of the largest sets on Earth. The sheer number of vowels is one reason that Danish is so hard to pronounce. Danish has 32 vowels, 15 short, 13 long and four unstressed: ɑ, ɑː, a, æ, æː, ɛ, ɛː, e, e̝ː, i, , o, , ɔ, ɔː, u, , ø, øː, œ, œː, ɶ, ɶː, y, , ʌ, ɒ, ɒː, ə, ɐ, ɪ, and ʊ.

There is also a strange phonetic element called a stød, which is a very short pause slightly before the vowel(s) in a word. This element is very hard for foreigners to get right.

Just about any word has at least four meanings, and can serve as noun, verb, adjective or adverb. Danish has two genders (feminine and masculine have merged into common gender), and whether a noun is common or neuter is almost impossible to predict and simply must be memorized.

Suggesting that Danish may be harder to learn than Swedish or Norwegian, it’s said that Danish children speak later than Swedish or Norwegian children. One study comparing Danish children to Croatian tots found that the Croat children had learned over twice as many words by 15 months as the Danes. According to the study:

The University of Southern Denmark study shows that at 15 months, the average Danish toddler has mastered just 80 words, whereas a Croatian tot of the same age has a vocabulary of up to 200 terms.

[…] According to the study, the primary reason Danish children lag behind in language comprehension is because single words are difficult to extract from Danish’s slurring together of words in sentences. Danish is also one of the languages with the most vowel sounds, which leads to a ‘mushier’ pronunciation of words in everyday conversation.

Danish gets a 3 rating, average difficulty.

Swedish has the disadvantage of having hundreds of irregular verbs. Swedish also has some difficult phonemes, especially vowels, since Swedish has nine vowels, not including diphthongs. Pronunciation of the ö and å (and sometimes ä, which has a different sound) can be difficult. Swedish also has pitch accent. Pronunciation is probably the hardest part of Swedish.

Words can take either an -en or an –ett ending, and there don’t seem to be any rules about which one to use. The same word can have a number of different meanings.

Swedish, like German, has gender, but Swedish gender is quite predictable by looking at the word, unlike German, where deciding which of the three genders to use seems like a spin of the Roulette wheel.

Word order is comparatively free in that one can write a single sentence multiple ways while changing the meaning somewhat. So I didn’t know that. can be written the following ways:

Det visste jag inte.
Det visste inte jag.
Jag visste inte det.
Jag visste det inte.
Inte visste jag det.

For some reason, Swedish got a very high score on a study of the weirdest languages on Earth.

The different ways of writing that sentence depend on context. In particular, the meaning varies in terms of topic and focus.

There is a 3-way contrast in deixis:

den här
den där

Swedish also has the same problematic phrasal verbs that English does:

att slå -  beat/hit

slå av     turn off
slå fast   settle/establish
slå igen   close/shut
slå igenom become known/be a success
slå in     wrap in, come true
slå ner    beat down
slå på     turn on
slå runt   overturn
slå till   hit/strike/slap, strike a deal
slå upp    open (a book), look s.t. up

Swedish orthography is difficult in learning how to write it, since the spelling seems illogical, like in English. The sj sound in particular can be spelled many different ways. However, Swedish spelling is probably easier than English since Swedish lacks a phonemic schwa, and schwa is the source of many of the problems in English. Where allophonic schwa does appear, it seems to be predictable.

One nice thing about Swedish grammar is that it is similar to English grammar in many ways.

Swedish can be compared to a tube in terms of ease of learning. The basics are harder to learn than in English, but instead of getting more difficult as one progresses as in English, the difficulty of Swedish stays more or less the same from basics to the most complicated. But learning to speak Swedish is easy enough compared to other languages.

Swedish gets a 2.5 rating, easy to average difficulty.


Any Gaelic language is tough. Celtic languages are harder to learn than German or Russian.

Insular Celtic

Old Irish was the version of Irish written from 650 to 900 AD. It was used only by the educated and aristocratic elites. The rest of the population spoke a simplified version that was already on its way to becoming Middle Irish.

The verbal system in Old Irish was one of most complicated of all of the classical languages.

The persons were 1st, 2nd, 3rd and plural. The tenses were present, preterite, imperfect, perfect, future and an odd tense called secondary future. There were imperative and subjunctive moods. There was no infinitive – instead it was formed rather erratically as a verbal noun derived from the verb. This gerund underwent 10 different declensions and often looked little like the verb it is derived from.

cingidto step -> céimstepping

There were both strong and weak verbs, and each had both simple and compound forms.

Bizarrely, every verb had not one but two different paradigms – the conjunct and the absolute. You used the conjunct when the verb is preceded by a conjunct particle such as (not) or in (the question particle). You used the absolute when there was no conjunct particle in front of the verb.

Hence, the present indicative of glenaid (sticks fast), is:

Absolute   Conjunct

glenaim    :glenaim
glenai     :glenai
glenaid    :glen
glenmai    :glenam
glenthae   :glenaid
glenait    :glenat

The colon before the conjunct verbs indicates that a conjunct particle preceded the verb.

The phonological changes were some of the most complicated you could imagine. An attempt was made to orthographically portray all of these convoluted changes, but the orthography ended up a total mess.

Each consonant had four different values depending on where it was in the word and whether or not it was palatal. Hence, even though the 1st person absolute and conjunct look identical above (both are spelled glenaim), they were pronounced differently. The absolute was pronounced glyenum, and the conjunct was pronounced glyenuv.

The grammar was unbelievably complex, probably harder than Ancient Greek. There was even a non-IE substratum running underneath the language.

Old Irish gets a 5 rating, extremely difficult.

Irish students take Irish for 13 years, and some take French for five years. These students typically know French better than Irish. There are inflections for the inflections of the inflections, a convoluted aspiration system, and no words for yes or no. The system of initial consonant mutation is quite baffling. Noun declension is mystifying. Irish has irregular nouns, but there are not many of them:

the womanan bhean
the women
na mná

and there are only about 10 irregular verbs. There are dozens of different declension types for verbs. The various phonological gradations, lenitions and eclipses are not particularly regular. There are “slender” and “broad” variants of many of the consonants, and it is hard to tell the difference between them when you hear them. Many learners find the slender/broad consonants the hardest part of Irish. The orthography makes many lists of worst orthographies on Earth.

Irish gets a 4.5 ratings, very difficult.

Both Scots Gaelic and Irish Gaelic are written with non-phonetic spelling that is even more convoluted and irrational than English. This archaic spelling is in drastic need of revision, and it makes learners not want to learn the language. For instance, in Scots Gaelic, the word for taxi is tacsaidh, although the word is pronounced the same as the English word. There are simply too many unnecessary letters for too few sounds. Of the two, Scots Gaelic is harder due to many silent consonants.

Irish actually has rules for its convoluted spelling, and once you figure out the rules, it is fairly straightforward, as it is quite regular and it is actually rational in its own way. In addition, Irish recently underwent a spelling reform. The Irish spelling system does make sense in an odd way, as it marks things such as palatalization and velarization.

Scottish Gaelic and Manx have gone a long time with no spelling reforms.

Scottish Gaelic gets a 4.5 ratings, very difficult.

Manx is probably the worst Gaelic language of all in terms of its spelling since it has Gaelic spelling yet uses an orthography based on English which results in a crazy mix that makes many lists of worst scripts.

Manx gets a 4.5 rating, very difficult.

Common Byrthonic

Welsh is also very hard to learn, although Welsh has no case compared to Irish’s two cases. And Welsh has a mere five irregular verbs. The Byrthonic languages like Welsh and Breton are easier to learn than Gaelic languages like Irish and Scots Gaelic. One reason is because Welsh is written with a logical, phonetic alphabet. Welsh is also simpler grammar-wise, but things like initial consonant mutations can still seem pretty confusing and are difficult for the non-Celtic speaker to master and understand. Verbal declension is irregular.

caraf   I love
carwn   we love

cerais  I loved
carasom we loved

The problem above is that one cannot find any morpheme that means 1st person, 3rd person, or past tense in the examples. Even car- itself can change, and in connected speech often surfaces as gar-/ger-. And carwn can mean I was loving (imperfect) in addition to we love. There are no rules here, and you simply have to memorize the different forms.

Welsh gets a 4 rating, very hard to learn.

Breton is about in the same ballpark as Welsh. It has a flexible grammar, a logical orthography and only four irregular verbs.

On the other hand, there are very few language learning materials, and most of those available are only written in French.

Breton gets a 4 rating, very hard to learn.


Greek is a difficult language to learn, and it’s rated the second hardest language to learn by language professors. It’s easy to learn to speak simply, but it’s quite hard to get it down like a native. It’s the rare second language learner who attains native competence. Like English, the spelling doesn’t seem to make sense, and you have to memorize many words. Further, there is the unusual alphabet. However, the orthography is quite rational, about as good as that of Spanish. Whether or not Greek is an irregular language is controversial. It has that reputation, but some say it is not as irregular as it seems.

Greek has four cases: nominative, accusative, genitive and vocative (used when addressing someone). There are three genders: masculine, feminine and neuter. Nouns have several different declension patterns determined by the ending on the noun. Verb conjugations are about as complicated as in Romance. Greek does retain the odd aorist tense. In addition, it has the odd middle voice and optative mood. Greek syntax is quite complicated.

Greek gets a 5 rating, extremely difficult to learn.

Classic or Ancient Greek was worse, with a distinction between aspirated and unaspirated consonants, a pitch accent system and a truly convoluted, insanely irregular system of noun and verb inflection. It had a dual number in addition to singular and plural and a very difficult optative case. Irregular verbs had one of six different stem types. The grammar was one of the most complex of all languages, and the phonology and morphology were truly convoluted.

Ancient Greek is said to have had four different genitive cases, but it actually had four different uses of the genitive:

  1. Objective Genitive – “for obedience to faith”
  2. Subjective Genitive – “faith’s obedience” or faithful obedience
  3. Attributive Genitive – “obedience of faith”
  4. Genitive of Apposition – obedience, i.e. faith

How confusing!

Classic Greek gets a 5.5 rating, nearly hardest of all to learn.


An  obscure branch of Indo-European, Armenian, is very hard to learn. Armenian is a difficult language in terms of grammar and phonetics, not to mention the very odd alphabet. The orthography is very regular, however there are some irregularities. For instance:

գրել , written grel but spoken gərel (schwa removed in orthography)
խոսել, written xosel but spoken xosal  (a changed to e in orthography)

However, the alphabet itself presents many problems. Print and cursive can be very different, and upper case and lower case can also be quite different. Here are some pairs of letters in upper and lower case:

Ա ա
Յ յ
Փ փ

All in all, this means you have to memorize as many as four different shapes for each letter. However, the grammar is very regular.

In addition, many letters very closely resemble other letters, which makes it very easy to get them mixed up:

գ and զ
and է
and ղ
and ռ

There are voiced consonants and an alternation between aspirated and unaspirated unvoiced consonants, so some mix up the forms for b, p and , for instance. Nevertheless, there are many things about the grammar that seem odd compared to other IE languages. For instance, Armenian has agglutination, and that is a very strange feature for an IE language.

Part of the problem is that due to its location in the Caucasus, Armenian has absorbed influences from some of the wild nearly Caucasian languages. For instance, an extinct NE Caucasian Nakh language called Tsov is thought to have contributed to the Hurro-Ururtian substratum in Armenian. So in a sense when you learn Armenian, you are also learning a bit of Chechen at the same time. For some reason, Armenian scored very high on a weirdest languages survey.

People who have learned both Arabic and Armenian felt that Armenian was much easier, so Armenian seems to be much easier than Arabic.

Armenian is rated 4, very hard to learn.


Albanian is another obscure branch of Indo-European. Albanian nouns have two genders (masculine and feminine), five cases including the ablative, lost in all other IE. Both definite and indefinite articles are widely used, a plus for English speakers. Most inflections were lost, and whatever is left doesn’t even look very IE. The verbal system is complex, having eight tenses including two aorists and two futures, and several moods, including indicative, imperative, subjunctive, conjunctive, optative and admirative. The last three are odd cases for IE. The optative only exists in IE in Ancient Greek, Sanskrit and Manx. Oddly enough, there is no infinitive. Active and passive voices are used.

Similarly to Gaelic, Albanian is even harder to learn than either German or Russian. Albanian may be even harder to learn than Polish.

Albanian is rated 5,extremely difficult.


All Slavic languages have certain difficulties. For instance, the problematic perfect/imperfect tenses discussed below in Czech and Slovak are present in all of Slavic. The animate/inanimate noun class distinction is present in all of Slavic also. Slavic languages also add verb prefixes to verbs, completely changing the meaning of the verb and creating a new verb (see Italian above).

East Slavic

People are divided on the difficulty of Russian, but language teachers say it’s one of the hardest to learn. Even after a couple of years of study, some learners find it hard to speak even a simple sentence correctly.

It has six basic cases – nominative, accusative, dative, genitive, instrumental and prepositional – and analyses have suggested up to 10 other cases. The most common of the extra cases are locative, partitive and several forms of vocative. All of these extra cases either do not apply to all nouns (“incomplete” cases) or seem to be identical to an existing case. At any rate, the vocative is only used in archaic prose. And there is also a locative case, which is what the exceptions to the prepositional case are referred to. Russian has two genitive cases, the so-called Genitive 1 and Genitive 2. The first one is standard genitive and the second is the genitive-partitive (see above), which is now only used in archaic prose.

The grammar is fairly easy for a Slavic language. The problem comes with the variability in pronunciation. The adjectives and endings can be difficult. In addition, Russian has gender and lots of declensions. Like Lithuanian, almost everything in the language seems to decline. The adjectives change form if the nouns they describe have different endings. Adjectives also take case somehow.

Verbs have different forms depending on the pronouns that precede them. Russian has the same issues with perfective and imperfective forms as Polish does (see the Polish section below). There are dozens of different declension types for verbs and many verbs that are irregular and don’t fit into any of the declension types. In addition, there are many irregular nouns, syncretisms, and an aspectual system that is morphologically unpredictable.

Word order is pretty free. For instance, you can say:

I love you by saying

I love you.
You love I.
Love you I.
I you love.
Love I you.
You I love

Pronunciation is strange, with one vowel that is between an ü and i. Many consonants are odd, and every consonant has a palatalized counterpart, which will be difficult to speakers whose languages lack phonemic palatalized consonants. These are the soft and hard consonants that people talk about in Russian. The bl sound is probably the hardest to make, but the trilled r is also problematic.

Russian has several words that, bizarrely, are made up of only a single consonant:

s with, off of
to, towards
in, into
– subjunctive/conditional mood particle (would)
Z – emphatic particle

In addition, Russian has some very strange words that begin with a doubled consonant sound:


The orthography system is irregular, so there are quite a few silent letters and words that are pronounced differently than they are spelled.

Word Silent Letters Example
здн  [знпраздник
рдц  [рцсердце
лнц  [нцсолнце
стн  [снлестница
вств [ств]          чувство
жч   [щ]            мужчина
зч   [щ]            извозчик
сч   [щ]            счастье
чт   [штчто
чн   [шнконечно
тц   [ц]            вкратце
дц   [ц]            двадцать
тч   [ч]            лётчик
дч   [ч]            докладчик
тся  [цца]          учится
ться [цца]          учиться

Stress is quite difficult in Russian since it seems arbitrary and does not appear to follow obvious rules:

дóмаat home

One problem is that phonemic stress, not written out, changes the way the vowel is pronounced. For instance:

узнаюI’m finding out
I will find out

The two are written identically, so how you tell them apart in written Russian, I have no idea. However in speech you can tell one from the other because the two forms have different stress.

Russian also has vowel reduction that is not represented in the orthography. The combination of stress and vowel reduction means that even looking at a Russian word, you are not quite sure how to pronounce it.

Like German, Russian builds morphemes into larger words. Again like German, this is worse than it sounds since the rules are not so obvious. In addition, there is the strange Cyrillic alphabet, which is nevertheless easier than the Arabic or Chinese ones. Russian also uses prepositions to combine with verbs to form the nightmare of phrasal verbs, but whereas English puts the preposition after the verb, Russian puts it in front of the verb.

All of Slavic has a distinction between animate and inanimate nouns as a sort of a noun class. Russian takes it further and even has a distinction between animate and inanimate pronouns in the male gender:

dvoje muzhchin     two men
troje muzhchin     three men
chetvero muzhchin  four men
pyatero muzhchin   five men
shestero muzhchin  six men
semero muzhchin    seven men

Compare to:

dva duba      two oaks 
tri duba      three oaks 
chetyre duba  four oaks

However, Russian only has the animate/inanimate distinction in pronouns and not in nouns in general.

Like Polish below, you use different verbs depending if you are going somewhere on foot or other than on foot. Second there is a distinction between going somewhere with a goal in mind and going somewhere with no particular goal in mind. For instance, to go:

idti (by foot, specific endpoint)
xodit’ (by foot, no specific endpoint)
exat’ (by conveyance, specific endpoint)
ezdit’ (by conveyance, no specific endpoint)

The verb to carry also has four different forms with the same distinctions as above.

In addition, there are various prefixes you can put on a verb:

into                  v-
out of                vy-
towards               po-
away from             u-
up to the edge of     pod-
away from the edge of ot-
through               pro-
around                ob-

These prefixes look something like “verbal case.” You an add any of those prefixes to any of the going or carrying verbs above. Therefore, you can have:

poiti  –walk up to something
drive around with no goal
–  walk away from something with no goal in mind

The combination of paths and goals results in some very specific motion verbs.

Russian is harder to learn than English. We know this because Russian children take longer to learn their language than English speaking children do. The reason given was that Russian words tended to be longer, but there may be other reasons.

Russian has the advantage of having quite a bit of Romance and Greek loans for a Slavic language, but unfortunately, you will not typically hear these words in casual conversion. Russian also has no articles. English speakers will find this odd, but others regard it as a plus.

Russian is less difficult than Czech, Polish or Serbo-Croatian.

Russian gets a 4 rating, very hard to learn.

West Slavic
Czech and Slovak

Czech and Slovak are notoriously hard to learn; in fact, all Slavic languages are. Language professors rate the Slavic languages the third hardest to learn on Earth. Czech is in the Guinness Book of World Records as the hardest language to learn. Even the vast majority of Czechs never learn to speak their language correctly. They spend nine years in school studying Czech grammar, but some rules are learned only at university. Immigrants never seem to learn Czech well, however, there are a few foreigners who have learned Czech very well – say, three or fewer errors in a 30 minute monologue, so it is possible to learn Czech well even if it is not very common.

Writing Czech properly is even more difficult than speaking it correctly, so few Czechs write without errors. In fact, an astounding 1/3 of the population makes at least on grammatical or spelling mistake in every sentence they write! The younger generation is now even worse as far as this goes, as Czech language teaching for natives has become more lax in recent years and drills have become fewer. Nevertheless, the Czech and Slovak orthographies are very rational. There is nearly a 1-1 sound/symbol correspondence.

Even natives often mess up the conditional (would). The 3rd conditional (past conditional) has nearly gone out of modern Czech and has merged with the present conditional:

3rd conditional – If I “would have known” it, I would not have asked has merged with
2nd conditional – If I “would know” it, I would not ask.

This means conditional events in the present are no longer distinguished between those in the past, and the language is impoverished.

Native speakers also mix up a specific use of the gerund:


She looked at me smiling.
He walked along whistling.
He was in his bed reading a book.

This is easy to say in English, and the use of these forms is rather common. However, it is very hard to make those sentences in Czech, and possibly only 3% of the population can formulate those sentences properly. Instead, they break them up into two sentences:


She looked at me, and she smiled.
He was in his bed, and he was reading.

Czech is full of exceptions and exceptions to the exceptions. It is said that there are more exceptions than there are rules. Czech has seven cases in singular and seven more cases in plural for nouns, for a total of 59 different “modes” of declension. There are also words that swing back and forth between “modes.” Adjectives and pronouns also have seven cases in the singular and plural. Czech is one of the few languages that actually has two genitive cases – one more or less possessive and the other more or less partitive. There are six genders, three in the singular and three in the plural.

When you put all that together, each noun can decline in 59 different ways. Further, these 59 different types of nouns each have 14 different forms depending on case. Verbs also decline. The verbs have both perfective and imperfective and have 45 different conjugation patterns. Czech learners often confuse the perfect and imperfect verbs. Verbs of motion can also be quite tricky.

One of the problems with Czech is that not only nouns but also verbs take gender, but they only do so in the past tense. In addition, Czech has a complicated aspect system that is often quite irregular and simply must be memorized to be learned.

This conjugation is fairly regular:

viděl continuous past – he saw
punctual – once he suddenly saw
repetitive – he used to see (somebody/something) repeatedly

Others are less regular:

jedl continuous – he ate
snědl dojedl
he ate it all up
he ate a bit of it
he finished eating
repetitive – he used to eat repeatedly

Czech also has an evidential system. The particle prý is used to refer to hearsay evidence that you did not personally witness.

Prý je tam zima.
Someone said/People say it’s cold outside.

Truth is that almost every word in the language is subject to declension. The suffixes on nouns and verbs change all the time in strange ways.

There are some difficult consonants such as š, č, ť, ž, ľ, ď, dz, , ĺ and ŕ. It’s full of words that don’t seem to have vowels.

Entire Czech sentences can have extreme consonant clusters that appear to lack vowels:

Strč prst skrz krk.
Stick a finger through your neck.

Smrž pln skvrn zvlhl z mlh.
A morel full of spots welted from fogs…

Mlž pln skvrn zvh.

However, the letters r and l are considered “half-vowels” in Czech, so the sentences above are easier to pronounce than you might think.

The letters ř and r (Czech has contrasting alveolar trills) are hard to pronounce, and ř is often said to exist in no longer language, including other Slavic languages. It is only found in one other language on Earth –  the Papuan language Kobon, which pronounces it a bit differently. Even Czechs have a hard time making these sounds properly (especially the ř), and many L2 speakers never get them right. There is also a hard and soft i which is hard to figure out.

As with other Slavic languages like Russian, it has the added problem of fairly loose word order. In addition, there are significant differences between casual and formal speech where you use different forms for someone you are familiar with (are on a first name basis with) as opposed to someone you do not know well. In addition, females use different endings for the past tense than men do.

On the plus side, Czech stress, like that of Polish, is regular as the accent is always on the first syllable. But if you come from a language such as Spanish where the accent is typically on the second syllable, this might present an obstacle.

Czech gets a 5.5 rating, nearly hardest of all.

Slovak is closely related to Czech, and it is controversial which one is harder to learn. Slovak is definitely more archaic than Czech. Some say that Slovak is easier because it has a more regular grammar. Slovak has the additional problem is marking acute accents: á, é, í, ĺ, ó, ŕ, ú and ý. Slovak fortunately lacks the impossible Czech ř sound. Instead it has something called a “long r,” (ŕ) which is not very easy to make either. This is something like the er sound in English her.

Slovak, like Czech, has retained the vocative, but it almost extinct as it is restricted to only a few nouns. Like Polish and Sorbian, Slovak also has an animate/inanimate distinction in gender for plural nouns. So Slovak has five genders: masculine, feminine and neuter in the singular and animate and inanimate in the plural.

Some say that Slovak is even harder than Polish, and there may be a good case that Czech and Slovak are harder than Polish.

Slovak gets a 5.5 rating, nearly hardest of all.


Polish is similar to Czech and Slovak in having words that seem to have no vowels, but in Polish at least there are invisible vowels. That’s not so obviously the case with Czech. Nevertheless, try these sentences:

  1. Wszczebrzeszynie chrząszcz brzmi w trzcinie i Szczebrzeszyn z tego słynie.
  2. Wyindywidualizowaliśmy się z rozentuzjazmowanego tłumu.
  3. W Szczebrzeszynie chrząszcz brzmi w trzcinie.

I and y, s and z, je and ě alternate at the ends of some words, but the rules governing when to do this, if they exist, don’t seem sensible. The letter ť is very hard to pronounce. There are nasal vowels as in Portuguese. The ą, ć, ę, ł, ń, ó, sz, cz, dz, , sounds are hard for foreigners to make. There are sounds that it is even hard for native speakers to make as they require a lot tongue movements. A word such as szczescie is hard to Polish L2 speakers to pronounce. Polish written to spoken pronunciation makes little sense, as in English – h and ch are one sound – h, ó and u are the same sound, and u may form diphthongs where it sounds like ł, so u and ł can be the same sound in some cases.

The confusing distinction between h/ch has gone of most spoken Polish. Furthermore, there is a language committee, but like the French one, it is more concerned with preserving the history or the etymology of the word and less with spelling the word phonemically. Language committees don’t always do their jobs!

Polish orthography, while being regular, is very complex. Polish uses a Latin alphabet unlike most other Slavic languages which use a Cyrillic alphabet. The letters are: A Ą B C Ć D E Ę F G H I J K L Ł M N Ń O Ó Q P R S T U V W X  Y Z Ź Ż. Even Poles say that their orthography is very complicated.

Polish is even complex in terms of pronunciation. There are apparently rules for regarding comma use, but the rules are so complex that even native speakers can’t make sense of them.

Further, native speakers speak so fast it’s hard for non-natives to understand them. Due to the consonant-ridden nature of Polish, it is harder to pronounce than most Asian languages. Listening comprehension is made difficult by all of the sh and ch like sounds. Furthermore, since few foreigners learn Polish, Poles are not used to hearing their language mangled by second-language learners. Therefore, foreigners’ Polish will seldom be understood.

Polish grammar is said to be more difficult than Russian grammar. Polish has the following:

There are five different tenses: zaprzeszły, przeszły, teraźniejszy, przyszły prosty, and przyszły złozony.

There are seven different genders: masculine animate, masculine inanimate, feminine, and neuter in the singular and animate and inanimate in the plural. However, masculine animate and masculine inanimate and the plural genders are only distinguished in accusative. Masculine animate, masculine inanimate and neuter genders have similar declensions; only feminine gender differs significantly.

Masculine nouns have five patterns of declension, and feminine and neuter nouns have six different patterns of declension. Adjectives have two different declension patterns. Numbers have five different declension patterns: główne, porządkowe, zbiorowe, nieokreślone, and ułamkowe. There is a special pattern for nouns that are only plural.

There are seven different cases: nominative, genitive, dative, accusative, instrumental, locative, vocative. Only the genitive locative cases are irregular, the latter only in the singular. Verbs have nine different persons in their declensions: ja, ty, on, ona, ono, my, wy, oni, one. There are different conjugation patterns for men and women. There are 18 different conjugation patterns in the verb (11 main ones). There are five different polite forms: for a man, a woman, men, women and men and women combined.

There are four different participle forms, three of which inflect. Some of these are active and others are passive, but the whole system is incredibly complex. All of the participles decline like nouns, each gender adds its bit to each pattern which in turn change more according to tense.

Polish has seven cases, including the vocative which has gone out of most Slavic. The vocative is often said to be dying out, becoming less common or only used in formal situations, but the truth is that it is still commonly used.

In an informal situation, a Pole might be more like to use nominative rather than vocative:

Cześć Marek! (Nom.), rather than
Cześć Marku! (Voc.)

However, in a more formal situation, the vocative is still likely to be used:

Dzień dobry panie profesorze/doktorze! (Voc.). Dzień dobry pan profesor/doktor! (Nom.) would never be used, even in casual conversation.

Case declension is very irregular, unlike German. Polish consonant gradation is called oboczność (variation).

The genders of nouns cause the adjectives modifying them to inflect differently.

matka    mother (female gender)
ojciec   father (male gender)
dziecko  child (neuter gender)

Modifying Adjective
brzydkiugly ugly

brzydka matka     ugly mother
brzydki ojciec    ugly father
brzydkie dziecko  ugly child

brzydkie matki    ugly mothers
brzydcy ojcowie  ugly fathers
brzydkie dzieci   ugly children

Gender even effects verbs.

I ate (female speaker) Ja zjadłam
I ate (male speaker)   Ja zjadłem

There are two different forms of the verb kill depending on whether the 1st person singular and plural and 2nd person plural killers are males or females.

I killed     zabiłem/zabiłam
We killed    zabiliśmy/zabiłyśmy
They killed  zabili/zabiły

The perfective and imperfective tenses create a dense jungle of forms:

kupować - to buy

Singular  Simple Past         Imperfect
I (f.)    kupiłam             kupowałam
I (m.)    kupiłem             kupowałem
you (f.)  kupiłaś             kupowałaś
you (m.)  kupiłeś             kupowałeś
he        kupił               kupował
she       kupiła              kupowała
it        kupiło              kupowało

we (f.)   kupiłyśmy           kupowałyśmy
we (m.)   kupiliśmy           kupowaliśmy
you (f.)  kupiłyście          kupowałyście 
you (m.)  kupiliście          kupowaliście
they (f.) kupiły              kupowały
they (m.) kupili              kupowali

The verb above forms an incredible 28 different forms in the perfect and imperfect past tense alone.

The existence of the perfective and imperfective verbs themselves is the least of the problem. The problem is that each verb – perfective or imperfective – is in effect a separate verb altogether, instead of just being conjugated differently.

The verb to see has two completely different verbs in Polish:


WidziałemI saw (repeatedly in the past, like I saw the sun come up every morning).
ZobaczyłemI saw (only once; I saw the sun come up yesterday).

Some of these verbs are obviously related to each other:


But others are very different:


This is not a tense difference – the very verbs themselves are different! So for every verb in the language, you effectively have to learn two different verbs. The irregular forms may date from archaic Polish.

In addition, the future perfect and future imperfect often conjugate completely differently, though the past forms usually conjugate in the same way – note the -em endings above. There is no present perfect as in English, since in Polish the action must be completed, and you can’t be doing something at this precise moment and at the same time have just finished doing it. 95% of verbs have these maddening dual forms, but for 5% of verbs that lack a perfective version, you only have one form.

It’s often said that one of the advantages of Polish is that there are only three tenses, but this is not really case, as there are at least eight tenses:

Indicative         grac       to play
Present            gram       I play 
Past               gralem     I played
Conditional        gralbym    I would play
Future             będę grać  I will play
Continuous future  będę grał  I will be playing
Perfective future  bogram     I will have played*
Perf. conditional  pogralbym  I would have played

*Implies you will finish the action

There is also an aspectual distinction made when referring to the past. Different forms are used based on whether or not the action has been completed.

Whereas in English we use one word for go no matter what mode of transportation we are using to get from one place to another, in Polish, you use different verbs if you are going by foot, by car, by plane, by boat or by other means of transportation.

In addition, there is an animate-inanimate distinction in gender. Look at the following nouns:

hat      kapelusz
computer komputer
dog      pies
student  uczen

All are masculine gender, but computer and hat are inanimate, and student and dog are animate, so they inflect differently.

I see a new hatWidze nowy kapelusz
I see a new student
Widze nowego ucznia

Notice how the now- form changed.

In addition to completely irregular verbs, there are also irregular nouns in Polish:

człowiek -> ludzie

Let us look at pronouns. English has one word for the genitive case of the 1st person singular – my. In Polish, depending on the context, you can have the following 11 forms, and actually there are even more than 11:


Numerals can be complex. English has one word for the number 2 – two. Polish has 21 words for two, and  all of them are in common use.

dwa (nominative non-masculine personal male and neuter and non-masculine personal accusative)
dwaj (masculine personal nominative)
dwie (nominative and accusative female)
dwóch (genitive, locative and masculine personal accusative)
dwom (dative)
dwóm (dative)
dwu (alternative version sometimes used for instrumental, genitive, locative and dative)
dwoma (masculine instrumental)
dwiema (female instrumental)
dwoje (collective, nominative + accusative)
dwojga (collective, genitive)
dwojgu (collective, dative + locative)
dwójka (noun, nominative)
dwójkę (noun, accusative)
dwójki (noun, genitive)
dwójce (noun, dative and locative)
dwójką (noun, instrumental)
dwójko (vocative)
dwojgiem (collective, instrumental)

Polish also has the paucal form like Serbo-Croatian. It is the remains of the old dual. The paucal applies to impersonal masculine, feminine and neuter nouns but not to personal masculine nouns.

Personal Masculine

one boy     jeden chłopiec
two boys    dwóch chłopców
three boys  trzech chłopców
four boys   czterech chłopców
five boys   pięciu chłopców
six boys    sześciu chłopców
seven boys  siedmiu chłopców
eight boys  ośmiu chłopców

Impersonal Masculine

one dog     jeden pies
two dogs    dwa psy
three dogs  trzy psy
four dogs   cztery psy
five dogs   pięć psów
six dogs    sześć psów
seven dogs  siedem psów
eight dogs  osiem psów

In the above, two, three and four dogs is in the paucal (psy), while two, three or four men is not and is instead in the plural (chłopców)

A single noun can change in many ways and take many different forms. Compare przyjacielfriend

                             Singular         Plural
who is my friend             przyjaciel       przyjaciele
who is not my friend         przyjaciela      przyjaciół
friend who I give s.t. to    przyjacielowi    przyjaciołom
friend who I see             przyjaciela      przyjaciół
friend who I go with         z przyajcielem   z przyjaciółmi
friend who I dream of        o przyjacielu    o przyjaciołach
Oh my friend!                Przyajcielu!     Przyjaciele!

There are 12 different forms of the noun friend above.

Plurals change based on number. In English, the plural of telephone is telephones, whether you have two or 1,000 of them. In Polish, you use different words depending on how many telephones you have:

two, three or four telefony, but
five telefonów.

Sometimes, this radically changes the word, as in hands:

four ręce, but
five rąk.

There are also irregular diminutives such as

psiaczek  -> słoneczko

Polish seems like Lithuanian in the sense that almost every grammatical form seems to inflect in some way or other. Even conjunctions inflect in Polish.

In addition, like Serbo-Croatian, Polish can use multiple negation in a sentence. You can use up to five negatives in a perfectly grammatical sentence:

Nikt nikomu nigdy nic nie powiedział.
Nobody ever said anything to anyone

Like Russian, there are multiple different ways to say the same thing in Polish. However, the meaning changes subtly with these different word combinations, so you are not exactly saying the same thing with each change or word order. Nevertheless, this mess does not seem to be something that would be transparent to the Polish learner.

In English, you can say Ann has a cat, but you can’t mix the words up and mean the same thing. In Polish you can say Ann has a cat five different ways:

Ania ma kota.
Kota ma Ania.
Ma Ania kota.
Kota Ania ma.
Ma kota Ania.

The first one is the most common, but the other four can certainly be used. The truth that while the general meaning is the same in each sentence, the deep meaning changes with each sentence having a slightly different nuanced interpretation.

In addition, Polish has a wide variety of dialects, and a huge vocabulary. Although Polish grammar is said to be irregular, this is probably not true. It only gives the appearance of being irregular as there are so many different rules, but there is a method to the madness underneath it all. The rules themselves are so complex and numerous that it is hard to figure them all out.

Polish appears to be more difficult than Russian. For example, in Russian as in English, the 1st through 3rd person past tense forms are equivalent, whereas in Polish, they are each different:

          English   Russian     Polish

1st past  I went    ya pashou   ja poszedłem 
2nd past  you went  ty pashou   ty poszedłeś
3rd past  he went   on pashou   on poszedł

Even adult Poles make a lot of mistakes in speaking and writing Polish properly. However, most Poles are quite proud of their difficult language (though a few hate it) and even take pride in its difficult nature.

On the positive side, in Polish, the stress is fixed, there are no short or long vowels nor is there any vowel harmony, there are no tones and it uses a Latin alphabet.

Polish is one of the most difficult of the Slavic languages. Even Poles say it is very hard to learn. Most Poles do not learn to speak proper Polish until they are 16 years old! Although most Poles know how to speak proper Polish, they often use improper forms when speaking formally, not because they do not know how to speak correctly but simply because they feel like

It is harder than Russian and probably also harder than Czech, though this is controversial. There is a lot of controversy regarding which is harder, Czech or Polish.

Polish gets a 5 rating, extremely difficult.

South Slavic

It’s controversial whether Bulgarian is an easy or hard language to learn. The truth is that it may be the easiest Slavic language to learn, but all Slavic language  are hard. Though it is close to Russian, there are Russians who have been living there for 20 years and still can’t understand it well.

It has few cases compared to the rest of Slavic. There are three cases, but they are present only in pronouns. The only case in nouns is vocative. This is odd because most Slavic languages have either lost or are in the process of losing the vocative, and in Bulgarian it is the only case that has been retained. Compared to English, Bulgarian is well structured and straightforward with little irregularity. In addition, Bulgarian has more Romance (mostly French) and Greek borrowings than any other Slavic languages. Romance came in via the Vlahs who lived there before the Slavs moved in and Greek from the Byzantine period. In recent years, many English borrowings have also gone in.

Bulgarian has a suffixed general article that is not found in the rest of Slavic but is apparently an areal feature borrowed from Albanian. The stress rules are nightmarish, and it seems as if there are no rules.

Bulgarian has grammatical gender, with three genders – masculine, feminine and neuter. In addition, adjectives must agree with the gender of the noun they are modifying. In English, adjectives are invariable no matter what the noun is:

pretty man
pretty woman
pretty horse
pretty table

However, the Bulgarian alphabet is comparatively simple compared to other Slavic alphabets. Since 1945, it has only had 30 letters. Compare this to the 70 letters in Polish. There are only six vowels, and it has the easiest consonant clusters in Slavic. The orthography is very regular, with no odd spellings. The Cyrillic alphabet is different for those coming from a Latin alphabet and can present problems. For one thing, letters that look like English letters are pronounced in different ways:

В is pronounced v in Bulgarian
E is pronounced eh in Bulgarian
P is pronounced r in Bulgarian

There are a number of Bulgarian letters that look like nothing you have ever seen before: Ж, Я, Ь, Ю, Й, Щ, Ш, and Ч. Bulgarian handwriting varies to a great degree and the various styles are often difficult to map back onto the typewritten letters that they represent.

While Bulgarian has the advantage of lacking much case, Bulgarian verbs are quite complex even compared to other Slavic languages. Each Bulgarian verb can have up to 3,000 forms as it changes across person, number, voice, aspect, mood, tense and gender. Bulgarian has two aspects (perfect and imperfect), voice, nine tenses, five moods and six non infinitival verbal forms.

For instance, each verb has at two aspects – simple and continuous – for each of the tenses, which are formed in different ways. Onto this they add a variety of derivatives such as prefixes, suffixes, etc. that change the meaning in subtle ways:

Aorist or Perfect:

да прочитамto read in whole a single text/book/etc (viewed as fact, that is the duration of the action does not interest us)
да изчитам – to read every book there is on the subject (viewed as fact, that is the duration of the action does not interest us)
да дочетаto finish reading something (viewed as fact, that is the duration of the action does not interest us)

Continuous or Imperfect:

да четаto be reading (viewed as an action in progress)
да прочитамto read in whole a single text/book/etc (viewed as an action in progress)
да изчитамto read every book there is on the subject (viewed as an action in progress)

Mood is very complicated. There are different ways to say the same idea depending on how you know of the event. If you know about it historically, you mark the sentence with a particular mood. If you doubt the event, you mark with another mood.

If you know it historically but doubt it, you use yet another mood. And there are more than that. These forms were apparently borrowed from Turkish. These forms are rare in world languages. One is Yamana, a Patagonian language that has only one speaker left.

In Bulgarian, you always know if something is a noun, a verb or an adjective due to its marking. You will never have the same word as an adjective, noun and verb. In English, you can have words that act as verbs, adjectives and nouns.

Let’s dance!
Let’s go to the dance.
Let’s go to dance lessons.

Bulgarian is probably the easiest Slavic language to learn.

Bulgarian gets a 3.5 rating, above average difficulty.

Macedonian is very close to Bulgarian, and some say it is a dialect of Bulgarian. However, I believe that is a separate language closely related to Bulgarian. Macedonian is said the be the easiest Slavic language to learn, easier than Bulgarian. This is because it is easier to pronounce than Bulgarian. Like Bulgarian, Macedonian has lost most all of its case. But there are very few language learning materials for Macedonian.

Macedonian gets a 3.5 rating, above average difficulty.


Serbo-Croatian, similar to Czech, has seven cases in the singular and seven in the plural, plus there are several different declensions. The vocative is still going strong in Serbo-Croatian (S-C), as in Polish, Ukrainian and Bulgarian. There 15 different types of declensions: seven tenses, three genders, three genres or moods, and two aspects. Whereas English has one word for the number 2 – two, Serbo-Croatian has 17 words or forms.

Case abbreviations below:
N = NAV – nominative, accusative, vocative
G = Genitive
D = Dative
L =Locative
I = Instrumental

Masculine inanimate gender
N dva
G dvaju
D L I dvama

Feminine gender
N dve
G dveju
D L I dvema

Mixed gender
N dvoje
G dvoga
D L I dvoma

Masculine animate gender
N dvojica
G dvojice
D L dvojici
I dvojicom

N dvojka
G dvojke
D L dvojci
I dvojkom

The grammar is incredibly complex. There are imperfective and perfective verbs, but when you try to figure out how to build one from the other, it seems irregular. This is the hardest part of Serbo-Croatian grammar, and foreigners not familiar with other Slavic tongues usually never get it right.

Serbian has a strange form called the “paucal.” It is the remains of the old dual, and it also exists in Polish and Russian.  The paucal is a verbal number like singular, plural and dual. It is used with the numbers dva (2), tri (3), četiri (4) and oba/obadva (both) and also with any number that contains 2, 3 or 4 (22, 102, 1032).

gledalac            viewer
pažljiv(i)          careful
gledalac pažljiv(i) careful viewer

1 careful viewer  jedan pažljivi gledalac 
2 careful viewers dva pažljiva gledaoca   
3 careful viewers tri pažljiva gledaoca   
5 careful viewers pet pažljivih gledalaca

Above, pažljivi gledalac is singular, pažljivih gledalaca is plural and pažljiva gledaoca is paucal.

As in English, there are many different ways to say the same thing. Pronouns are so rarely used that some learners are surprised that they exist, since pronimalization is marked on the verb as person and number. Word order is almost free or at least seems arbitrary, similar to Russian.

Serbo-Croatian, like Lithuanian, has pitch accent – low-rising, low-falling, short-rising and short-falling. It’s not the same as tone, but it’s similar. In addition to the pitch accent differentiating words, you also have an accented syllable somewhere in the word, which as in English, is unmarked. And when the word conjugates or declines, the pitch accent can jump around in the word to another syllable and even changes its type in ways that do not seem transparent. It’s almost impossible for foreigners to get this pitch-accent right.

The “hard” ch sound is written č, while the “soft” ch sound is written ć. It has syllabic r and l. Long consonant clusters are permitted. See this sentence:

Na vrh brda vrba mrda.

However, in many of these consonant clusters, a schwa is present between consonants in speech, though it is not written out.

S-C, like Russian, has words that consist of only a single consonant:


Serbo-Croatian does benefit from a phonetic orthography.

It is said that few if any foreigners ever master Serbo-Croatian well. Similar to Czech and Polish, it is said that many native speakers make mistakes in S-C even after decades of speaking it, especially in pitch accent.

Serbo-Croatian is often considered to be one of the hardest languages on Earth to learn. It is harder than Russian but not as hard as Polish.

Serbo-Croatian gets a 4.5 rating, very difficult.

Slovenian or Slovene is also a very hard language to learn, probably on a par with Serbo-Croatian. It has three number distinctions, singular, dual and plural. It’s the only major IE European language that has retained the dual. Sorbian has also retained the dual, but it is a minor tongue. However, the dual may be going out in Slovenia. In Primorska it is not used at all, and in the rest of Slovenia, the feminine dual is not used in casual speech (plural is used instead), but the masculine dual is still used for masculine nouns and mixed pairs of masculine and feminine nouns.

In addition, there are six cases, as Slovene has lost the vocative. There are 18 different declensions of the word son, but five of them are identical, so there are really only 13 different forms.

   Singular Dual       Plural 
1. Sin      Sina       Sini
2. Sina     Sinov      Sinov
3. Sinu     Sinovoma   Sinovom
4. Sina     Sinova     Sinove
5. O sinu   O sinovoma O sinovih
6. S sinom  Z sinovoma Z sini

There are seven different ways that nouns decline depending on gender, but there are exceptions to all of the gender rules. The use of particles such as pa is largely idiomatic. In addition, there is a lack of language learning materials for Slovene.

Some sounds are problematic. Learners have a hard time with the č and ž sounds. There are also “open” and “closed” vowels as in Portuguese.

Here is an example of a word that can be difficult to pronounce:


However, Slovene has the past perfect that is the same as the English tense, lost in the rest of Slavic. In addition, via contact with German and Italian, many Germanic and Romance loans have gone in. If you know some German and have some knowledge of another Slavic language, Slovene is not overwhelmingly difficult.

Some people worry that Slovene might go extinct in the near future, as it is spoken by only 2 million people. However, even this small language has 356, 881 headwords in an online dictionary. So it is clear that Slovene has plenty enough vocabulary to deal with the modern world.

Slovene is easier than Russian, Serbo-Croatian, Polish, Czech or Slovak.

Slovenian gets a 4 rating, very hard.


Eastern Baltic

Lithuanian, an archaic Indo-European Baltic tongue, is extremely difficult to learn. There are many dialects, which is interesting for such a small country, and the grammar is very difficult, with many rules. There is grammatical gender for nouns, and in addition, even numerals have gender in all cases. The language is heavily inflectional such that you can almost speak without using prepositions.

A single verb has 16 participial forms, and that is just using masculine gender for the participles. You can also add feminine forms to that verb. There are two main genders or giminės, masculine and feminine, but there is also neutral gender (bevardė giminė), which has three different forms. Verbs further decline via number (singular, dual and plural) and six different cases. There are five classes of verbs and six modes of declension for nouns (linksniai). However, Lithuanian verb tense is quite regular. You only need to remember infinitive, 3rd person present and 3rd person past, and after that, all of the conjugations are regular.

Here is an example of the Lithuanian verb:

Eiti – “to go. Ei is the verb root, and ti is in infinitival suffix.

Verbs decline according to:

Person and number
1st singular einu   I go  
3rd dual     einava we two go
1st plural   einame we go

The four tenses

2nd pl. past       Ėjote    you (guys) went
2 sing. imperfect  eidavote you used to go
2 sing. indicative einate   you go
2 sing. future     eisite   you will go

They also change according to something called “participants.” The participant paradigm has three tenses and all three genders. Participants are further divided into direct and indirect.

Regular direct participant (3 tenses, 3 genders)

Ėjęs   while he himself went
einąs  while he himself is going
eisiąs while he himself will be going

Ėjusi  while she herself went

buvo einama while it itself went
einama      while it itself was going
bus einama  while it itself will be going

Regular indirect participant (3 tenses, 3 genders)

past    eidytas     one that was forced to go
present eidomas     one that is being forced to go
future  bus eidomas one that will be forced to go

Semi participant (no tenses, 2 genders)

eidamas while going himself

eidama  while going herself

Active participant (2 tenses, no genders)

past    Ėjus   while going (in the past)
present einant while going now

2nd infinitive or budinys (no tenses)

eite in a way of going

Plusquamperfect (be + regular participants)

indicative būti   to have been gone
present    yra    has been gone
past       buvo   had been gone 
imperfect  būdavo used to have been gone 
future     bus    will have been gone

past 3pl   buvo ėję they had been gone 

Additional moods 

Imperative (all persons) 

Eik!             Go! 
Eikime!          Let's go! 
Teeina/Lai eina! Let him/her go! 

Subjunctive (all persons) 
eičiau I would go 
eitum  thou would go

In addition, while most verb marking is done via suffixes, Lithuanian can make aspect via both suffixes and prefixes, bizarrely enough (Arkadiev 2011).

Determining whether a noun is masculine or feminine is easier than in German where you often have to memorize which noun takes which gender. Lithuanian is similar to Spanish in that the ending will often give you a hint about which gender the noun takes.

Here is an example of the sort of convolutions you have to go through to attach the adjective good to a noun.

geras - good

             Masculine          Feminine

             Singular  Plural   Singular  Plural
Nominative   geras     geri     gera      geros
Genitive     gero      gerų     geros     gerų
Dative       geram     geriems  gerai     geroms
Accusative   gerą      gerus    gerą      geras
Instrumental geru      gerais   gera      geromis
Locative     gerame    geruose  geroje    gerose

The noun system in general of Lithuanian is probably more complicated even than the complex Russian noun system. Lithuanian is possibly more irregular and may have more declensions than even Polish. Learners often feel that the grammar is illogical.

Furthermore, while it does not have lexical tone per se, it does have pitch accent – there are three different pitches or degrees (laipsniai), which sound like tones but are not tones. Stress is hardly predictable and nearly needs to be learned word by word. It’s almost impossible for foreigners to get the accent right, and the accents tend to move around a lot across words during declension/conjugation such that the rules are opaque if they exist at all. It was formerly thought to be nearly random, but it has now been found that Lithuanian stress actually falls into four paradigms, so there is a system there after all.

You cannot really forget about lexical tone when learning Lithuanian, as stress is as fundamental to Lithuanian as tone is to Mandarin.

Often you need a dictionary to figure out where the accent should be on a word. Lithuanian pronunciation is also difficult. For example, look at rimti (to get calm) and rimti (serious – plural, masculine, nominative). There is a short i sound that is the same in both words, but the only difference is where the stress or pitch accent goes. Consonants undergo some complicated changes due to palatalization. Lithuanian has soft and hard (palatalized and nonpalatalized) consonants as in Russian.

Try these words and phrases:

šąla šiandien
ačiū už skanią vakarienę
čežėti šiauduose

Or this paragraph:

Labas, kaip šiandien sekasi? Aš esu iš Lietuvos, kur gyvenu visą savo gyvenimą. Lietuvių kalba yra sunkiausia iš visų pasaulyje. Ačiū už dėmesį.

Lithuanian is an archaic IE language that has preserved a lot of forms that the others have lost.

In spite of all of that, picking up the basics of Lithuanian may be easier than it seems, and while foreigners usually never get the pitch-accent down, the actual rules are fairly sensible. Nevertheless, many learners never figure out these rules and to them, there seem to be no rules for pitch accent.

Learning Lithuanian is similar to learning Latin. If you’ve been able to learn Latin, Lithuanian should not be too hard. Also, Lithuanian is very phonetic; words are pronounced how they are spelled.

Some languages that are similar to English, like Norwegian and Dutch, can be learned to a certain extent simply by learning words and ignoring grammar. I know Spanish and have been able to learn a fair amount of Portuguese, French and Italian without learning a bit of grammar in any of them.

Lithuanian won’t work that way because due to case, base words change form all the time, so it will seem like you are always running into new words, when it fact it’s the same base word declining in various case forms. There’s no shortcut with Latin and Lithuanian. You need to learn the case grammar first, or little of it will make sense.

Some say that Lithuanian is even harder to learn than the hardest Slavic languages like Polish and Czech. It may be true.

Lithuanian gets a 5 rating, extremely hard to learn.

Latvian is another Baltic language that is somewhat similar to Lithuanian. It’s also hard to learn. Try this:

Sveiki, esmu no Latvijas, un mūsu valoda ir skanīga, skaista un ar ļoti sarežģītu gramatisko sistēmu.

Latvian and Lithuanian are definitely harder to learn than Russian. They both have aspects like in Russian but have more cases than Russian, plus a lot more irregular verbs. Latvian, like Lithuanian, has a tremendous amount of inflection. The long vowels can be hard to pronounce.

Latvian is easier to learn than Lithuanian. The grammar is easier to figure out and the phonological system is much easier. Also, Latvian has lost many archaic IE features that Lithuanian has retained. Latvian has regular stress, always on the first syllable, as opposed to Lithuanian’s truly insane stress system. Latvian has fewer noun declensions, and fewer difficult consonant clusters.

Latvian gets a 4.5 rating, very hard.


Arkadiev, Peter. 2011. On the Aspectual Uses of the Prefix Be- in Lithuanian.
Baltic Linguistics 2:37-78.
Seymour, Philip H. K.; Aro, Mikko; Erskine, Jane M. and the COST Action A8 Network. 2003. Foundation Literacy Acquisition in European Orthographies. British Journal of Psychology 94:143–174.

This research takes a lot of time, and I do not get paid anything for it. If you think this website is valuable to you, please consider a a contribution to support more of this valuable research.


Filed under Albanian, Applied, Armenian, Baltic, Bulgarian language, Celtic, Czech, Danish, Descriptive, Dutch, English language, French, Gaelic, German, Germanic, Greek, Hellenic, Hindi, Icelandic, Indic, Indo-European, Indo-Hittite, Irish Gaelic, Italian, Italic, Kashmiri, Language Families, Language Learning, Language Samples, Latvian, Linguistics, Lithuanian, Norwegian, Polish, Portuguese, Romance, Russian, Sanskrit, Serbo-Croatian, Sinhala, Slavic, Slovak, Spanish, Swedish