This is a post I have been wanting to write for awhile now. Kinyarwanda words often look fairly easy to pronounce the way they are written, but if you are unfamiliar with the language, just trust me: you will say them wrong. Rwandans will still appreciate your effort, but they might have to strain to understand you. Moreover, I have seen other written efforts to explain pronunciation, and outside of linguistic studies everyone does it wrong.
I am not myself Rwandan, but I have studied both Kinyarwanda and linguistics for the past two years now, and I feel like the pronunciation is one thing that I have got down pretty well. Perhaps I am flattering myself by thinking that this will be useful to students or expats, but if any were to find it I think it would be.
My apologies to readers who enjoy reading posts about what I am doing in Rwanda; the major thing that I did today was write this pronunciation guide! Feel free to skip or skim it, or maybe just read the conclusion. Unless you are actually interested in Kinyarwanda, you might find it pretty boring.
Vowels
Kinyarwanda has five vowels: a, e, i, o and u. They are pronounced approximately as they would be in Castilian (standard) Spanish. In American English terms:
- a is like the a in “father,” or the o in Standard American English (henceforth S.A.E.) “hot.” It is not pronounced like in “hat” or “hate.”
- e is a lot like the ea in S.A.E. “pear.” It is not like the ee in “meet.” It is also not like the a in S.A.E. “hate”; it is similar, but listen closely and you will notice that that vowel actually ends with an ee-ish sound. You want only the first part; it is closer to the e in S.A.E. “pet,” and sometimes does sound like that, but your best bet is imitating that vowel in “pear.”
- i is pronounced like the ee in “meet.” It also sometimes sounds like the i in “pit.” It is not pronounced like the i in “mine.”
- o is a lot like the o in S.A.E. “pore.” It is not like the o in “pot.” It is also not like the oa in S.A.E. “coat”; it is close, but listen closely and you will notice that that vowel actually ends with an oo-ish sound. You want only the first part; it is closer to the aw in Northeastern U.S. English “paw,” and sometimes does sound like that, but your best bet is imitating that vowel in “pore.”
- u is pronounced like the oo in S.A.E. “pool.” It also sometimes sounds like the u in “put.” It is not pronounced like the u in “cut” or, if it sounds different to you, the ew in “new.” If you are from California, you probably have an unusual u; find a friend from somewhere else and imitate the way they say it.
- y is always a consonant and never a vowel.
In linguistic terms, Kinyarwanda has a very standard five-vowel system, consisting of sounds that the International Phonetic Alphabet (henceforth I.P.A.) represents as [i], [e], [ɑ], [o] and [u].
It is also useful for English-speakers to note that Kinyarwanda is written in such a way that there is no such thing as a silent vowel: every vowel is pronounced the same way everywhere. Vowels can also be short or long, or have high or low tones; these are not written, which is annoying. More on that later, if you are interested.
Consonants
I will try to break these down in a way that still makes sense to someone not familiar with linguistics (though I will categorize them using linguistic terms):
Voiced Stops*
- b is not like English b, though Rwandans will understand if you pronounce it that way. Really, it is a voiced bilabial fricative, I.P.A. [β]. In terms of sound, this sounds like a mixture between a b, a w and a v. If that is not enough, make a v sound, and notice that you pronounce it with your bottom lip against your upper teeth; to make a [β], make the same sound but with both lips rather than lips and teeth. (It might not be easy.) Note that after m, or before w or y, b is pronounced exactly as in English.
- d is pronounced like in English.
- jy has a range of different pronunciations, depending on where you are in the country. The easiest for English-speakers to pronounce is like the j in “juice,” I.P.A. [d͡ʒ]. The most standard, however, is harder: a voiced palatal stop, I.P.A. [ʝ]. To imitate this without taking a phonetics course, try pronouncing a hard g with a y. Say “gya” a few times and notice that where your tongue touches your mouth; then move that spot forward, so that you are kind of flattening your tongue against the roof of your mouth (the palate) and then releasing; you could also try it by starting with a j + y and moving backward. (This is also hard.)
- g is pronounced like a hard English g (as in “go”), usually. When followed by a, o or u, it always is. When followed by e or i, pronunciation varies; in some regions, people still pronounce it like in “go,” but not usually. The most common sound, I think, under these circumstances, is [ʝ] as described above, though you will also hear a hard English j and everything in between.
- m and n are pronounced as expected, as is ny. (It might be worth noting here that ny, I.P.A. [ɲ], is pronounced with your tongue against the roof of your mouth, unlike n; you probably do this without thinking.)
* Kinyarwanda b is not actually a stop, but rather a fricative. This looks like a recent development, though, and the consonant inventory as a whole still makes more sense if it is considered a stop.
Voiceless Stops
- p is pronounced like in English “pot,” I.P.A. [pʰ]. (In the north, you will hear people say it like English b, but don’t worry too much about that.)
- t is pronounced like in English “top,” I.P.A. [tʰ].
- cy is like jy, except voiceless—that is, a voiceless palatal fricative, I.P.A. [cʰ]. Follow the instructions for jy above, except use English ch instead of j, and k instead of g.
- k is pronounced like English k, usually (I.P.A. [kʰ]), but shows the same alternations as g. In other words, it is sometimes pronounced like cy or c before e and i. See instructions for g above.
Note that, when they follow m or n, all of these sounds change and basically sound like h; see the “diphthongs” section below.
Affricates
- pf is basically pronounced like a p with an f; more accurately, it is like German pf, as in Kampf, I.P.A. [p͡ɸʰ].
- ts is basically pronounced like a t with an s, I.P.A. [t͡sʰ], though it may be useful for you to think of it like German z, as it is used to begin syllables.
- c is pronounced like the ch in English “chop,” I.P.A. [t͡ʃʰ]. It is not pronounced like the c’s in “cot” or “brace.”
Note that, when they follow m or n, these sounds might change; see the “diphthongs” section below.
Fricatives
- f and v are pronounced as in English.
- s and z are pronounced as in English. (Note that s is always pronounced as in “sock,” never as in “raise.”)
- sh is pronounced as in English.
- j is pronounced like the s in “treasure.” It is not pronounced like the j in “joke.”
- shy is kind of hard. It is a voiceless palatal fricative, I.P.A. [ç]. To get it, try saying shya a few times, then hya a few times, and then find the common place in the middle. You should end with your tongue almost flat against your palate: shy is pronounced in the same place as cy and jy. (If you do it right, you might hear a bit of a whistle.)
- h is pronounced as in English.
Approximants and Unused Letters
- w is pronounced as in English.
- r is tapped, like the r in Spanish pero, I.P.A. [ɾ]. If you have trouble pronouncing this (as many people do), try saying a phrase like “prince of Prussia” or “How the Grinch Stole Christmas,” but replace your r’s with d’s (i.e. “pdince of Pdussia,” etc.) and say them really fast a few times. You might notice that the d seems to get weaker, and eventually you are just tapping your tongue against your mouth. That is the sound you want. Kinyarwanda r is not pronounced like English r; if you cannot get the tap, you will be understood with your native r, you’ll just sound like a foreigner. A speaker might occasionally trill his r, like in Spanish perro (I.P.A. [r]), but this is nonstandard.
- y is always a consonant and never a vowel. It is pronounced as in English “young.”
l is not used in standard Kinyarwanda. Under older writing standards, r was written as l before i and e, because someone thought the sound was different, but it generally is not. In some cases, most notably Kigali, the old spelling of a place-name was considered too entrenched to change, and has been kept. Also, in loanwords, l is still written, e.g. ubupolitiki “politics,” and pilipili “chili-pepper” (from Swahili), though it is pronounced like r. If you see l under other circumstances, however, it is nonstandard and should, in any case, be pronounced as r.
q and x are not used in Kinyarwanda.
Diphthongs
Prenasalized Sounds
Prenasalized sounds are consonants preceded by m or n.
- mb is pronounced as in English “thimble,” not as in “thumb” and not with a normal Kinyarwanda b (see above).
- nd, njy and ng are pronounced as you would expect: the nasal consonant before the stop described above. Note that ng is pronounced as in “single,” not as in “sing.”
- mp, nt and nk (ncy does not occur) are not pronounced as you would expect. The standard pronunciation sounds as though the second sound in each is turned to h, though you might still hear a shadow of it. (Phonetically, my interpretation is that these are actually nasally released stops, though this is debatable.) So they are, respectively, pronounced kind of like mh, nh and ngh. Sometimes, you may even hear the h drop, and the first sound lengthen to compensate, but that is nonstandard.
- mpf, nts and nc are often pronounced as expected, given the descriptions above. Regionally, however, many people drop the middle sound, leaving mf, ns and nsh, respectively, which are pronounced intuitively.
- mf, mv, ns, nz, nsh, nj and nshy are pronounced as expected, given the descriptions above. (nh does not occur.)
- See subsequent sections for mw, nw, nyw and nny. mm, nn and nr do not occur.
You will often see words beginning with a nasal consonant cluster (e.g. nshaka, mpa). These are not typos, and there is no vowel between them or before them. Those two example words are considered to have two and one syllables, respectively. If you have trouble, try holding out the first consonant (e.g. mmmmmmm) and then pronouncing the rest of the word; say it a few times, shortening the first sound each time until it is not longer than what comes after it.
Labialized Sounds
Labialized sounds are consonants followed by w.
- bw is pronounced as bg. It’s weird (as is everything in this section and the next), but that is what happens. Try to pronounce them kind of at the same time; there should be a very small gap, but really not much. (vw does not occur.)
- dw, zw, jw and rw are pronounced with a g between the two sounds, so dgw, zgw, jgw and rgw.
- gw is pronounced as expected; jyw, if it ever occurs, is very rare.
- pw and fw are pronounced as pk and fk. Some people pronounce fw as just f, and pfw does not occur.
- tw, sw, tsw, cw and hw are pronounced with a k between the two sounds, so tkw, skw, tskw, ckw and hkw.
- kw is pronounced as expected; cyw and shyw are rare and hard to pronounce, but if they occur they would be pronounced as expected.
- mw is pronounced as mng (I.P.A. [mŋ]). umwana is kind of pronounced um-nga-na, with ng being the last consonant in “thing.”
- nw is pronounced as expected.
- nyw is pronounced as ngw, again with ng as in “thing,” so kunywa is kind of pronounced as kung-wa.
Consonant clusters that are both prenasalized and labialized, e.g. ntw and mbw, are mostly pronounced by adding a nasal consonant before the above pronunciations. For sounds in this category with voiceless stops (in this case p, t and k), also pronounce the middle sound as an h. So nkw is pronounced as nghw—thus, inkweto is kind of pronounced as ing-hwe-to.
Palatalized Sounds
Labialized sounds are consonants followed by y.
- Palatalized voiced sounds (by, dy, vy and ry) are made by inserting a jy (or, if you prefer, a g) between the two sounds. They come out sounding like bjy, djy, vjy and rjy, with each component sound pronounced as discussed above. ry is probably the hardest Kinyarwanda sound for a foreigner to produce, and it may be best to just hear it from a native speaker. ry and dy often sound the same, and sometimes even both come out as jy, but neither of these is standard. gy and zy do not occur.
- Palatalized voiceless sounds (py, ty and sy) are made by inserting a cy (or, if you prefer, a k) between the two sounds. They come out sounding like pcy, tcy and scy, with each component sound pronounced as discussed above. ky does not occur.
- my is basically pronounced my, though it is really more like mny.
- nny is really unusual: it is an n, followed after a very brief pause by ny. So kunnya is kind of pronounced ku-n(e)-nya, though it can sound like ku-n(e)ng-ya also. This sound is very hard to produce and also not common.
Consonant clusters that are both prenasalized and palatalized, e.g. nty and mby, are mostly pronounced by adding a nasal consonant before the above pronunciations.
Unwritten Things
Kinyarwanda’s orthography (writing system) is actually pretty good: it mostly reflects the way people think of sounds in their heads, and there is perfect consistency in representation: pronunciations of consonants, for example, do not change unpredictably as they do in languages like English (where spelling makes much more sense in terms of what things sounded like several centuries ago).
It is not, however, ideal. The problem with it is that Kinyarwanda vowels can be long or short, and also either high or low; these distinctions are critically important to being able to speak the language, and yet the standard orthography does not mark them at all. The only real way to understand and properly produce these is to make a point of learning about the language (difficult) or to get a good dictionary (probably harder—to my knowledge, all dictionaries in print are pretty bad).
It is possible, however, to make some educated guesses, as there are patterns present. In the following sections, I will briefly describe tone and length in Kinyarwanda, and give some tips on how a foreigner can try to imitate and predict them. For starters, note that I will write words a slightly different way here than the standard: long vowels will be doubled (e.g. aa), and vowels with high tones will have acute accents (e.g. á). I’ll explain more as I go along.
Length
In English, we have a weird concept of vowel length: we think of the a in “hate” as long, and the a in “hat” as short, even though they take about the same amount of time to pronounce. Moreover, phonetically, those two vowels are not really even related. (This is because older English did actually have vowels with different lengths, but then the long ones turned into different vowels and eventually shortened. Look up the “Great English Vowel Shift” for more.)
Anyway, to the extent that you can, scrap that idea. In Kinyarwanda—and, frankly, in almost every language where it is important—a long vowel is one that is held out for a longer time than its short counterpart. The two are otherwise identical. In fast, spoken Kinyarwanda, the difference is sometimes diminished, but it is still usually audible.
In Kinyarwanda, length is an important way of distinguishing words: urutoke (all short) means “finger,” but urutooke (with a long o) means “banana-plantation.” Those two words, however, are both written as urutoke. A word written as gusura might be gusuura “to visit,” or it might be gusura “to fart.” If you speak Kinyarwanda natively, it is not hard to distinguish them based on context, but for learners it is maddening.
In cases like the two above, in which basic forms of words are identical except for length, the only way of knowing the difference is to know each specific case. In the rest of the language, however—where, even if it is not the only important feature, the length of the vowel is about as important as whether it is an e or an o—there are some patterns that can make it a little bit easier:
- The last syllable of a word is always short.
- The first syllable of a word, if it begins with a vowel, is always short.
- A syllable that begins with a consonant and either y or w is always long (unless it is the last syllable of a word). Thus, syllables beginning with shw, kw, bw, py, ry, etc. will be long. (Note that ny and shy are thought of as units, and therefore this rule does not apply to them; syllables starting with these sounds might be long, but you cannot know for sure.)
- A syllable is always long if the next syllable begins with m or n and another consonant. (I phrase it that way because in Kinyarwanda every syllable ends with a vowel—thus impwempwe “chest-hair” is divided as i.mpwe.mpwe.) Thus, vowels coming before mb, nkw, nd, nshy, mpy, etc. will be long unless they are the first sound in the word. (Here again, the rule does not apply to ny, which is thought of as one sound.)
- In general, most syllables are short. When in doubt, just assume they all are and wait for someone to correct you.
- Certain suffixes recur frequently, and are always either long or short. For example, many verbs end will -iisha or -eesha, which is long (though written short); similarly, -ira, -era, -ora, -ura, -ika and -eka as verbal suffixes are short, as is the common prefix -ra-/-da (unless the fourth rule here applies to it). Get to know these and you will have an easier time.
Tones
If you are not familiar with linguistic tone, here is a quick summary. Many languages in the world, especially concentrated in East Asia, sub-Saharan Africa and South America, use tone—like higher and lower musical pitch—as a meaningful part of their speech. If you only speak English or another European language, this is hard to wrap your mind around and especially hard to train yourself to hear. Think of how saying “no” in English with different intonation can communicate kind of different moods: you can convey despair (all high), curiosity (middle rising), disbelief (high falling), sarcasm (low falling), resignedness (all low) and others. This is like that, except a lot more explicit: like, a word that sounds the same to an American might have completely different meanings depending on its tone. For example, in Kinyarwanda, inda means “stomach,” but índa (with a high first syllable) means “louse.” umuryaango means “family,” but umúryáango means “doorway.”
Some languages have lots of different tones: standard Mandarin has five, and Cantonese has eight. Kinyarwanda is simpler in that respect—it only has two, high and low—but it is more complex in that the tones are a critical part of grammar (verb tenses, clause distinctions, etc.) as well as distinguishing words.
The rules governing tones are pretty complicated; here is a brief and simplified summary:
- The default tone is low. High tones are common, but form a minority of syllables.
- There is an important, predictable difference between the way the tones of a word appear abstractly (in a person’s head) and the way they come out of the mouth. This is a general linguistic idea that applies to other things also: we always think of the plural marker in English as -s, but without thinking we pronounce it as -z in words like “words” and “ribs.” The next few rules describe this difference as it applies to tones.
- High tones “anticipate”; that is, if people think of a word as having a high tone on one vowel, they say it with a high tone on the previous vowel as well. So isakú “high tone” is pronounced as isákú; umuryáango “doorway” is pronounced as umúryáango; daatá “my father” is pronounced as daátá; and igikóokó “large animal” is pronounced as igíkóókó.
- There is an exception to the previous rule: if a high tone comes on the second half of a long vowel, it is not anticipated. Therefore, ubutiínde “vowel-length” is pronounced as ubutiínde, not ubutíínde.
- There is a tone-falling rule also, by which the original high tone is actually pronounced as a short falling tone, kind of to form a bridge between high and low. Here, I will indicate falling tone with a circumflex, e.g. â. There is an exception, though: if a high tone appears on the first half of a long vowel, it does not fall. Therefore, the examples above are actually pronounced as isákû, umúryáango, daátâ, igíkóókô and ubutiînde.
- There are more rules about tone-spreading and -neutralization that are very complicated and interact with grammar and morphology in such a way that they are far beyond the scope of this guide, and also not necessary unless you really want to learn the language in detail and from a linguistic perspective.
Tones are very hard to hear in conversation. They are incorporated into the overall melody of the sentence that contains them, and especially on short syllables they are almost inaudible in fast speech. On one level, this is really annoying and makes learning them exceedingly difficult. On the other hand, it also means that the short tones are not as important to imitate properly, especially as a beginner or well-meaning expat.
Long tones, however, are usually still audible, even in normal speech if you listen hard. Given the rules above, it is possible for long syllables to have five different tone patterns:
- They can be low and flat, as in ishyaamba “forest.” A low, flat tone is at the baseline pitch of your speech, perhaps a little bit lower. Given the rules above, this kind of syllable may be followed by a high or low tone; it may be preceded by a low or falling tone.
- They can be high and flat, as in áméézâ “table.” This tone is somewhat, noticeably higher than the baseline, but not so much that you have to strain. Given the rules above, this kind of syllable must result from an “abstract” high tone on the first part of the long vowel, and also on the following syllable. Therefore, it is almost always preceded by a high tone (which is a little bit lower than the one on the long syllable), and almost always followed by a falling tone.
- They can be high and low (falling), as in úmwáana “child.” This results from an abstract high tone on the first part of the long vowel. Therefore, this kind of syllable always occurs after a high tone and before a low tone. The preceding high tone is not as high as the one on the long syllable, which starts at a peak and then slides downward for its duration, ending at the baseline.
- They can be low and high (rising), as in umúsôroórô, some kind of tree. This is probably the least common tone combination, for whatever reason, and results from an abstract high tone on the following syllable. It starts low and steadily rises until it peaks at the end. This kind of syllable may follow either a low or falling tone, and will almost always be followed by a falling tone.
- They can be low and then falling (peaking), as in umuhuûngu “boy.” This results from an abstract high tone on the second part of the long syllable, which subsequently falls. This kind of syllable starts low and rises a bit half way through, but falls before it reaches the level of a high tone. This syllable will usually appear between two low syllables, though the preceding vowel may be falling and the following vowel may be high.
Tones are very difficult, if even possible, to predict without knowing, as they can generally appear anywhere in a root. The exceptions are verbs, which may have an abstract high tone on the first vowel of their root (or not), but nowhere else. (These verbs may acquire other grammatical high tones to indicate tense and aspect, however.) The far past marker, -á-; the reflexive marker, -íi-; and the subjunctive marker, -é, have high tones. All noun-class prefixes and the standard subject and object verb affixes carry a low tone unless one is anticipated onto them from the following syllable. Verbs beginning with the negator nti-, or containing the future marker -zaa- or the recent past marker -a- will not have high tones anywhere. In reduplicated stems and nouns derived from verbs, there are also rules about where high tones may be placed, and in general the language has rules governing “tone rhythms.” These are complex, though, and require background knowledge of the language to be useful.
Having observed for awhile, I sometimes think I notice patterns: it seems, for example, like syllables beginning with mw, cy and a couple of other sounds are likely to have a high-low tone on their long vowel. It also seems like long vowels before ng or nk usually have low-falling tones. There are counter-examples to all of these perceived patterns, however, and in the end you probably just have to know the words.
Stress
One of the reasons tone and length are so difficult for English-speaking learners of Kinyarwanda—and other Bantu languages, I am sure—is that English and most other Indo-European languages have neither as important parts of their grammar. What those languages have instead is stress.
You probably have an idea of what stress (or accent) is: one syllable in each English word seems to be emphasized more than all the rest (“probably,” “idea,” “accent”…). Phonetically, stressed syllables are louder than others, but also—and this is why it’s confusing—tend to be longer, and have a higher pitch.
Generally, languages have either stress or tone. There are a couple of key differences between the two: (a) syllables can only be either stressed or unstressed, whereas there can be several different kinds of tones, and (b) one syllable in a stressed language is always more stressed than all the others, whereas a language like Kinyarwanda can have as many high tones as it wants in a single word (or none at all).
So Kinyarwanda does not have stress, at least not in a way that is important to its speakers. Tone and vowel-length are important in different ways, and it is hard for learners to get used to the fact that high tones and long vowels do not correlate with stress, nor are they accompanied by noticeably higher volume.
All that being said, some syllables are still pronounced more loudly than others (stressed, if you’d like). In general, though I have not made a study of it, the first syllable of the stem of a verb seems to come out louder, and otherwise usually the second-to-last syllable. It is not an important part of speech, however. My advice to learners would be to try to minimize stress as much as they can, then listen to native speakers talk and try to fit your speech into the rhythm you hear from them.
Normal Speech
Kinyarwanda, as spoken normally, is very hard to understand, for a variety of reasons. On one level, fluent speakers of this as any other language just speak really fast, and that is hard enough. Also, because of the kind of language Kinyarwanda is, it contains a lot of really long words with many small parts that are very difficult to parse as a learner.
Another feature that makes comprehension difficult is one that can be imitated as a learner. Kinyarwanda is a language that does not allow words to end in consonants, so it is ironic that the last vowel in a word is almost never fully articulated. In single words, a speaker will often start to say it very briefly, but swallow it or trail off. (Phonetically, I think they devoice it, making it all but inaudible.).
In full sentences, if one word ends with a vowel and the next begins with a vowel, only the second is pronounced (except in very careful speech). Vowels, and tones also, are elided together in such a way that the words come out very fluidly and, if you stop trying to understand and just listen, sound very nice.
Conclusion
Kinyarwanda is a hard language; it has many sounds and phonetic features that are very unusual to speakers of English and related languages, and that does not even take into account all of the complex grammar. I guess that is why this guide came out to be so much longer than I intended it to be.
Rwandans often ask me why I choose to study Kinyarwanda. They point out that it is a language only spoken by a few million people, and not used internationally in the way the English and French are. They point out that it is really hard, and will likely not be terribly useful professionally. These are both fair points. I usually explain that I study linguistics, and it helps me to study the structure of Kinyarwanda to see the concepts I learn about abstractly being put to concrete use in a real language. This is true. I actually started learning it on a whim, but then I liked it so much that I kept going with it.
The things I get out of studying and speaking this language, though, are often less tangible and mostly unrelated. First, Rwandans are legitimately surprised and pleased to see a foreigner making an effort to learn their language; they are willing to tolerate my mistakes and they encourage me, and that is really rewarding. Second, I like it as a mental exercise: understanding the workings of another language, learning about the rules that govern it and the history that brought it to where it is now, and challenging myself to find ways to say new things are all really stimulating. Puzzling through these things is one of the better ways I can find to make my mind grow, often in unexpected ways.
On top of that, Kinyarwanda is a beautiful language, and knowing about it brings with it a wealth of cultural knowledge and sometimes a new way of thinking. Both of these are inestimably valuable, and I would not have had access to them any other way.
What I have just written does not only apply to this language: two years of studying linguistics has taught me that every language is equally intricate and complex, and is often intimately attached to the culture of its speakers. But this is a response to those who question the use of learning a “minor” language: in a nutshell, it is the experience of learning that I think provides the most benefit, and the fact that people do not expect my knowledge of it makes their gratitude that much more rewarding.
So, if you are an expatriate living in Rwanda or even a tourist just passing through, I would encourage you to try to learn a little bit of the mother tongue of your adopted home. If, for whatever reason, you find yourself as a guest in a place where the language spoken is not your own, give it a try, out of appreciation for your hosts. Even if you have never left your home country in your whole life and never plan to, I am very much in favor of language-learning as a way to strengthen your brain and expand your horizons. And I might add that this one is pretty cool.
