Guidelines for Translators
This is a set of guidelines intended to help translators using SuttaCentral’s Pootle. It is advisory only, nothing is mandatory.
Tools: Pootle, PO, and HTML are your friends!
The software you’re reading this on is called Pootle. Pootle is a free and open source program for translation, which is web-based. It’s likely that you are used to using word processors and the like for translation, so take a little time to get used to Pootle.
Pootle offers a variety of tools to help translators. In addition, we’ve added our own specialized tools, such as Pali lookup and terminology. These will help you translate better and faster.
Pootle works with a kind of file called PO. We create PO files from the texts on SuttaCentral, which are in HTML. HTML is the basis of the internet, and so it is the most widely used document type. Our developers have written code that converts our HTML to PO, and then, when the translation is finished, back to HTML again. That means that you never have to worry about the files you’re working on. Just do the translation and everything will work. In addition, because the HTML used on SuttaCentral is well-structured, it is easy to use on other websites, or to convert into other forms if you wish.
Here’s an example of what our PO code looks like behind the scenes. Don’t worry, you never have to use this, it’s purely for your interest.
msgctxt "an4.123:4.1" msgid "“Cattārome, bhikkhave, puggalā santo saṃvijjamānā lokasmiṃ." msgstr "“Mendicants, these four people are found in the world."
As you can see, there’s three elements: the source text (
msgid), the translated text (
msgstr), and the ID (
msgctxt). Since each translated segment has a unique ID, which is consistent across all translations, we can match source text to translated text everywhere, in all languages.
In addition to these elements, the PO file will record any HTML tags that are associated with that text, so that it can be restored when needed.
The segmenting of the text is done by our programs, and is based simply on punctuation in the source text. Since our Pali text is generally quite consistent, this gives fairly good results. However, this is not always the case, which explains why sometimes similar or identical passages are segmented differently.
Generally speaking, each segment should contain a source and a translation of that source. However, this is not crucial. In some cases it’s convenient to change the order of the translated strings (for example in verses). Or you might not translate everything (for example in highly repetitive texts). That’s fine. The only crucial thing is that the translation reads well in your language. Everything else is secondary. The point is that the matching is useful, so you can find the source text for any translation. Even if it doesn’t match exactly, you can still find it close by, so it’s not that big of a deal.
Style: there are older and simpler and better words
The essence of translation is invisibility. Disappear, and let the content shine. The Dhamma doesn’t need your cleverness. Your readers are distracted, stressed, and busy, and have carved out a small chunk of their day to read a Sutta. Be kind to them by making their job as simple as possible.
Once Hemingway was criticized by William Faulkner for his simple vocabulary, and responded:
Poor Faulkner. Does he really think big emotions come from big words? He thinks I don’t know the ten-dollar words. I know them all right. But there are older and simpler and better words, and those are the ones I use.
It is obvious that a simpler text will be preferred by non-native speakers, people with little education, or with reading disabilities. However, studies also show that highly educated readers strongly prefer simpler texts. A simple text is less annoying, and doesn’t make readers feel stupid for not understanding.
Readability can be measured by a range of widely-used tests. We’ve run these comparing our translation with that of Bhikkhu Bodhi, which you can check here. The results show that our text is significantly more readable.
The UK Government website has a useful summary and set of links for further reading.
- Use short sentences: Research strongly shows that shorter sentences are easier to read. Pali commonly uses very long sentences, and where possible they should be broken into shorter segments. Compare, for example, the second part of AN 4.147. BB’s translation is one sentence of 86 words, while mine is five sentences averaging 17 words. Another example is AN 4.200, where BB has one sentence of 163 words (!), while I have 3 sentences averaging 24 words. Here’s a tip I follow: each time you use a semi-colon, ask if it could be done with a full stop.
- Convey the essence of the sentence at the start: Readers skim. Give them what they need at the start of the sentence, leaving elaborations for the later part of the sentence, especially in cases where it is not possible to break the sentence into smaller pieces. (See what I did there?)
- Bear in mind George Orwell’s famous 6 points for good writing:
- Never use a metaphor, simile, or other figure of speech which you are used to seeing in print.
- Never use a long word where a short one will do.
- If it is possible to cut a word out, always cut it out.
- Never use the passive where you can use the active. (Note: the Pali itself sometimes shifts from passive to active in parallel passages, showing it is often of no great significance.)
- Never use a foreign phrase, a scientific word, or a jargon word if you can think of an everyday English equivalent.
- Break any of these rules sooner than saying anything outright barbarous.
- Phrase things in a positive way: Pali has a strong preference for negative and double-negative forms. These require more cognitive effort to process, especially when they are not idiomatic for your language. Rephrase in a positive way where possible. There are some contexts where the negative form is philosophically important, and of course these should be kept. However there are also many cases where a negative form must mean the positive inverse, not mere negation. AN 4.200 is a good example of this. Some English forms do this: “disliking” is not mere absence of liking.
- The best meaning is the least meaning: We tend to read a lot of significance into Dharma texts, sometimes taking little details, coincidences, or literary parallels as having a great significance, and bending the translation to try to capture these things. But mostly we’re just reading shapes in the clouds; and even if the implication is genuine, still the primary thing is the basic intent of the passage at hand. That must be clear, whatever else. Hence the best rendering is often the one that has least meaning. See here for an interesting discussion of this principle in the context of Bible translation.
- Don’t translate out of fear: Fear of being wrong, fear of what others will say, or fear that people will be put off. Sometimes Dhamma is disturbing: even the devas thought so. Don’t make the text off-putting; we should generally do as the Buddha did and fill readers with joy. But don’t hide the fact that Nibbana means “extinguishment”, or that monks in the old days were “mendicants”.
There’s little to be gained by slavishly imitating oral features of the text. I’ve chosen to mostly omit the repeated “bhikkhave” found in almost every sentence, except at the beginning of a text. I also omit honorifics after the first use in a text.
When it comes to handling repetitions, this is the general policy I have followed:
- Reduce the number of abbreviations that refer outside the present sutta
- Abbreviate as much as reasonable within the present sutta
The idea is to minimize the amount of times a reader has to refer elsewhere to complete a passage. At the same time, we don’t make the reader wade through unnecessary repetions within a text.
The guiding consideration behind this policy is the fact that our text will be on the web. In this environment, it is no longer read from one sutta to the next, but is essentially “random access”. If reading a book, we can assume that someone reads one sutta, then reads the next, so they can remember what that passage said. Of course, even in books this is a big assumption and doesn’t always work, but still, it is true to some extent. Online it is still true to some extent, but less so. This is why I try to minimize having to leap from one sutta to the next. However, the practical reality is that in a general edition this cannot be totally eliminated. There are simply too many repetitions
One of the reasons I’ve done these things is because, unlike most translators, I have been as much a part of the oral tradition as I have the written. I’ve been a patimokkha reciter for over twenty years, and in my youth I memorized dozens of suttas. I could recite Pali non-stop, from memory, all night if I wanted. Alas! It’s mostly forgotten now, though I still keep up the patimokkha.
After many hundreds of hours reciting Pali, it becomes obvious that the very same features that are most annoying in a written text often have exactly the opposite effect in recitation. The repetitions, the vocatives, all these things create a soothing rhythm. They give space and ease to the text, making it more enjoyable to recite. They also help to create “non-thinking” space, where you can contemplate the meaning as you recite.
But when reading, these same features become annoying. They’re a cognitive grit, constantly requiring filtering out. They encourage shallow skim reading, as you assume a text is not worth lingering over. While the linguistic content can be mirrored by imitating these conventions, the emotional response is exactly the opposite. And, as AN 5.26 teaches us, it is through a positive emotional response that a text becomes liberating.
If we look to Buddhist tradition, we see that after the adoption of writing, written texts discarded these techniques, and from that time the early texts became little used. People prefer to read something that’s readable. This is one of the major factors behind the lack of interest and knowledge of the early texts in Buddhism down to the present day. If, through a sense of linguistic purity, we insist on keeping these features, we ghettoize the Suttas, restricting them to a tiny group of specialists.
For me the main point is not whether someone will end up with a better understanding of the Dhamma by reading my translation or by reading another translation. It’s whether they’ll get a better understanding of the Dhamma by reading any translation or by never reading any suttas at all. Because that is how it is now in pretty much every Buddhist community. Creating a readable translation is one step towards changing that.
Take care to use punctuation accurately and consistently. Sloppy punctuation is a red flag for a careless text. SuttaCentral’s punctuation style for English is presented here. Be sure to adapt these where necessary for your language. If you need help with getting punctuation working properly—especially auto-corrected quote marks—let us know.
- Single space between sentences.
- Use correct dashes:
- Hypen (-) connects words, making word-combinations.
- En-dash (–) indicates a range, usually of numbers (AN 4.45–56).
- Em-dash (—) sets apart a clause—like this—in the middle of a sentence. This is particularly useful in dealing with long sentences in Pali that are not amenable to splitting into smaller segments. Use the em-dash to set apart non-essential or subsidiary elements of a sentence. The sentence should still make sense if you skip the part in em-dashes. This can be very helpful in presenting a more clearly articulated sentence.If you can’t easily input these, use two hyphens for en-dash, and three hyphens for em-dash. Don’t worry what it looks like, we’ll convert it later. Be consistent! Also, notice that none of these dashes takes a space on either side.
- Apart from the case of setting apart clauses as mentioned above—which only applies to em-dash—use of colon and em-dash is very similar. Here’s a rule of thumb: prefer the colon, except when setting apart clauses—whether at the end of sentences of the beginning. However—as you can see in this paragraph—overuse of strong punctuation—of any kind—is annoying and bad style; avoid it where possible in favor of breaking sentences into shorter, more coherent ideas.
- Punctuation typically takes a space after, not before (quotes and dashes excepted).
- Use correct quote marks. Our editor will autocorrect "straight quotes" to “typographically correct quotes”. However, quotes vary a lot from language to language, so make sure it works for you. Never use crappy internet pseudo quotes!
- Ensure proper nesting of quote marks. For English our convention is “…‘…“…”…’…”.
Use Logical Quotation. This is not followed in the Pali text, but we shouldn’t follow its bad example. Eg:
kā ca pana vo antarākathā vippakatā”ti?
Even though the question is part of the Buddha’s words, and thus the question mark logically falls inside the quote marks, it is placed outside. This is bad! Do this instead:
What conversation was interrupted?”
The EBTs have a lot of extended passages in quotation. The normal convention is that when a quotation extends over multiple paragraphs, you place opening quotes on each paragraph in the sequence, and closing quotes only on the last paragraph. This is used in Ven Bodhi's editions. Another option, used in the Numata editions, is to set the quoted passages as blockquotes. This is very tempting, especially for online texts; but when the majority of the text is blockquotes, it creates wasteful margins in printed editions. Our Pali text follows a simpler convention, placing open quotes at the start and close quotes at the end, regardless of whether there are any paragraph breaks. I have followed this style. One advantage of this is that it is more robust; consider what happens if, for example, someone decides to do paragraphing differently. While there is some possibility of inviting confusion, since practically all the doctrinal text is quotation anyway, I don’t think this will be a problem.
- Use proper ellipsis (…), never three dots (...). This should autocorrect for you. Typically we have a space either side.
- Use the Oxford comma.
I recommend avoiding the academic practice of indicating supplied terms in translations with [square brackets]. These are unscientific, misleading, and ugly. They detract from the text and convey a misleading sense of false precision. Consider just one example, Ven Bodhi’s translation of SN 27.3. Here the Pali text begins to abbreviate (with no variants that I know of). The Pali has Yato kho, bhikkhave, bhikkhuno … pe … abhiññā sacchikaraṇīyesu dhammesū”ti. Literally,
When, mendicants, a mendicant … in regard to things realizable by direct knowledge.
Ven Bodhi has
When a bhikkhu has abandoned the mental corruptions in these six cases … [his mind] becomes wieldy in regard to those things that are to be realized by direct knowledge.
Obviously the extra text has been added to make the abbreviated portion readable, which is fine. But why is [his mind] singled out as supplied text? It seems completely arbitrary. Thus the use of square brackets here, and commonly elsewhere, merely creates false precision. The reality is that we are adding or removing things in every sentence, and so we should. * Not really punctuation, but for convenience, here is a list of words that are not capitalized in headings: a, an, the, at, by, for, in, of, on, to, up, and, as, but, or, nor.
Use Commentaries and Parallels with Discretion
Sometimes you can get useful insights from the commentaries and parallels. However the text before you must be the final arbiter.
The traditional Pali commentaries are a two-edged sword. They’re indispensible in countless cases, yet after a hundred years we are still translating the commentarial interpretation too often. In many cases, explanations based on the commentary actively confuse the situation, introducing complex and unnecessary ideas, and making it harder to understand the text. Given the hold the commentaries have had for 2,000 years, and their often unconscious influence on our current understandings, it would be understandable to simply translate against the commentaries whenever possible. If the text has more than one interpretation, and one agrees with the commentary, use the other one.
This is, however, a crude approach. Much more useful is to be aware of the commentary’s theoretical postions, and be on the lookout for how these are influencing its explanations. If you haven’t done so already, I recommend a close study of the Visuddhimagga. This will give you a good grasp on the main commentarial approaches, and you can recognize them when they are read back into the text.
Here’s one example. It’s only a small point, but an interesting one. In AN 4.127 and elsewhere, there’s a memorable description of deep space: lokantarikā aghā asaṃvutā. BB translates asaṃvutā as “abysmal”, which has suitably awesome ring to it, but is a curious choice. Asaṃvuta, more well known in the sense “unrestrained”, stems from the old root √vṛ, famous in the Rig Veda for the serpent Vṛtra who wraps the world in darkness.
How do we get from “wrapped, bounded” to “abysmal”? The commentary says Asaṃvutāti heṭṭhāpi appatiṭṭhā (‘Unbounded’ means, not established even below.) This explanation also made its way into the PTS Dictionary: “ungoverned orderless, not supported, baseless”. Now, something that has no bottom can be said to be an “abyss”, and something abyss-like is “abysmal”. So it seems like a clever rendering.
But note how it relies on not one, but two stretches: we have to stretch the Pali from “unbounded” to “bottomless (abyss)” via the comms. But it’s another stretch in English from abyss to abysmal. In fact in modern English the underlying metaphor of abysmal is virtually gone, and it just means “extremely bad”; “abyss-like” is an archaic or literary sense. So this is already dubious.
The Chinese doesn’t have any such sense of “badness” here, translating literally as 無有障蔽, i.e. without bounds. Likewise, the Critical Pali Dictionary says “not closed; with no material limits”.
However, we need to also consider what the purpose of the commentary’s explanation is. In his note on the text, BB presents without comment the standard commentarial cosmology, which depicts the world systems arranged in two dimensions. Rather than the three dimensional galaxies we’d expect, the tradition holds that the world systems are like plates arranged next to one another on a table. Needless to say, there’s no hint of this two-dimensional cosmology in the Suttas.
But it makes a problem for interpreting this passage: in the two dimensional model, interstellar space is in fact bounded by the solar systems on either side. So how can it be “unbounded”? This must be why the commentary stresses that the space is unbounded “below”. So the whole idea of the “abyss” really stems from an attempt by the commentary to buttress its interpretation—one which, needless to say, is wrong—which ended up influencing both PTS Dictionary and translation.
I rendered the phrase as “boundless desolation of interstellar space”. Just for fun I googled it and found the sentence: “Innumerable small molecules plunge quietly through the boundless, desolate regions of interstellar space.” That’s exactly what I’m aiming for, to render the meaning so that it sounds like something a modern native speaker might actually use. If Google tells you that a phrase is only found in Buddhist texts, it’s time to question your choices.
It’s a translation. Only leave words in an Indic form when absolutely necessary.
Keeping words in their Pali (or Sanskrit) form is by no means a guarantee that the meaning will be retained accurately. Words are not fixed. Their meaning is determined by usage, and in most cases Pali words are used today differently than in the Suttas. For example:
- Bodhisattva: Google says it means: “a person who is able to reach nirvana but delays doing so through compassion for suffering beings.” This idea is not only absent from the Suttas, but contradicts virtually everything they say on the topic.
- Karma: Today usually means “good or bad luck, viewed as resulting from one’s actions.” It should be avoided unless the context makes it clear that it means deed.
- Nirvana: Often means an ill-defined state of eternal bliss. Use with care, if at all.
- Samādhi: In Buddhist cultures means “meditation”, in Hinduism, “union with the divine”.
Only retain Indic terms if you’re confident that your readers will understand them correctly.
I have translated more fully than BB’s translations, in such areas as the following.
- Flora & Fauna: Relying on Dhammika’s dictionary (part of SC). Wherever possible I give the common English name for plants, etc.
- Celestial phenomena: Rather than use “ten-thousand-fold world system” I use “galaxy”, and so on.
- Numbers: I use best guesses for large numbers. The use of modern Indian numbers like lakh is particularly unhelpful.
- Measurements: Yojana is actually quite close to a league, and so I translate it and other measurements where possible.
- Social classes: Most of the ancient Indian social classes have fairly close parallels in Western societies. Khattiyas are “nobles” or “aristocrats”, but noble is used for ariya. The only one that’s really untranslatable is brahmin, as there is no Western notion of a hereditary priesthood.
- Vinaya offenses: In order, I use: expulsion, suspension, undetermined, relinquishment with redemption, redemption, confession, training, settling of issues, grave offense, wrong-doing.
Note that I haven’t translated the uddānas at the end of each vagga. That’s because I think to do so is a waste of time. Essentially they were just the Table of Contents of their day. Nowadays they’re only really relevant in working out things like relationships between manuscripts, for which you need to use the Pali anyway. However if you do want to translate them, feel free. They’ll be marked up just fine when we produce the HTML.
Incidentally, in the few cases of words that I retain in Indic I usually use the Sanskrit as it is better known, though I personally dislike it.
Here’s a list (incomplete!) of terms that I have left in the Indic, leaving aside proper names.
- Specialized religious terms: Veda, brahmin (semi-Indic!), uposatha
- papra (a tree)
- Buddha, Dharma, Saṅgha
Prefer gender-neutral language, except where gender is, or may be, relevant to the text. In some cases this is, or is seen as, less elegant style. But I have no qualms in adjusting style if it means being more inclusive. Idioms such as the generic you or the singular they have a long history of being used by great writers and condemned by grammatical purists. In recent years the notion of gender and its relation to language has become even more progressive, so much so that “they” was declared “word of the year” by the American Dialect Society. You might even argue that such idioms fit better with the colloquial style of Pali, rejecting the formalism of Sanskrit and moving towards a more inclusive mode of expression, in contrast with the Vedic texts, which were, in theory at least, restricted to male brahmins.
If you believe that it is essential to preserve the gendered pronouns, consider this. Pali is a gendered language. Every noun is masculine, neuter, or feminine. But English is not like this: we only have genders for the pronouns. So no matter what you do, you’ll end up removing maybe 95% of the gendered information in the text. In fact, every single translator has done this, and no-one even noticed. We do this even when the noun refers to an actual gendered entity (like a cat). Consider, also, how this works in various languages. In French, you’ll have to gender your nouns, and they won’t agree with the genders in Pali. In Thai, most pronouns aren’t gendered. In Chinese, they’re pronounced ungendered, but spelled with gender—but only due to modern Western influence. And so on.
The Chinese case is particularly interesting. In modern oral Chinglish—English spoken as a second language by native Chinese speakers—it’s common to mix up or just ignore the gender of pronouns. As European languages in the past changed Chinese, perhaps Chinese is now changing English. In 2000, an estimated 390 million Chinese had studied English. If even a small percentage of these achieve reasonable fluency, it is still one of the world’s most significant communities of English speakers, and the number is still increasing, especially among the young. It’s not far fetched to think that the loss of the gendered pronoun in Chinglish, together with its rapid decline in the West, will result in the non-gendered forms becoming standard in a very short time. Texts written with gendered pronouns will become linguistic orphans, sounding as old-fashioned to future ears as the old PTS translations sound to ours.
And it’s not as if the use of gender in Pali itself is unproblematic. As just one example, two of the words referring to women—mātugāma (woman, womenfolk) and dāra (wife)—are in fact declined as masculine. Are we to use a masculine pronoun with these?
In any translation you will have to deal with gender in a manner that’s specific to your language. But languages do not determine expression. There are many ways of saying the same thing. Languages make these different modes of expression possible. You can choose to express the Dhamma in a way that makes gender prominent, or you can choose to express it in a way that does not. Either way, you will never be able to duplicate the actual gendered details of the source text. Language just doesn’t work that way. Rather than trying, and inevitably failing, to mimic meaningless grammatical details, is it not wiser and more compassionate to express the Dhamma in such as way as to include and embrace, to speak to all without marginalizing? Should we not have special care for women and those of non-binary gender, whose voices are so rarely heard in Dhamma texts?
The important question is whether it affects the meaning. I don’t think it’s appropriate to disguise gendered passages where gender is important. But the question we should ask, I think, is this: what do we gain by using a masculine viewpoint as default in places where gender is not a factor? In the vast majority of cases, I don’t think anything is gained at all. But I am sure something is lost.
In most cases suttas deal with common spiritual, ethical, and psychological questions, and there is no reason to think that gender plays any meaningful role. They could be just as well addressed to men or women, or both. In fact we have many examples of this in the suttas themselves. For example, AN 4.170 begins by referring to “monks and nuns” but then continues with just “monks”. Obviously this is just a convention of abbreviation. Typically Pali refers to different groups or individuals in the vocative by just addressing the more senior, such as when Anuruddha stands in for his friends.
Another example is found in DN 16:
Yo kho, ānanda, bhikkhu vā bhikkhunī vā upāsako vā upāsikā vā dhammānudhammappaṭipanno viharati sāmīcippaṭipanno anudhammacārī, so tathāgataṃ sakkaroti garuṃ karoti māneti pūjeti apaciyati, paramāya pūjāya.
Any monk or nun or male or female lay follower who practices in line with the teachings, practicing properly, living in line with the teaching—they honor, respect, revere, venerate, and esteem the Realized One with the highest honor.
Here the stock demonstrative/relative pronouns (yo/so) are in masculine singular, but they explictly refer to men and women. While it is somewhat unusual to find cases that are so explicit, it is clear that the use of the masculine gender in Pali grammar is not always reserved for males.
Another example is found in DN 19 Mahāgovinda, where the stock formula for going forth “shaving hair and beard” is applied to a list that includes the brahmin’s wives and the harem women.
There are some obvious cases where the gender should be preserved, such as when it involves a specific person, or when it discusses gender.
More problematic are cases where the text implies or deals with topics that may not be inherently gendered, but which were gendered in that context.
For example we have the idiom purisathāmena purisavīriyena purisaparakkamena, “with manly strength, manly energy, and manly vigor”. It’s not necessary to translate this in a gendered way. You could say, for example, “heroic strength …” or “dauntless strength …”. However, the idiom clearly draws on an association of masculinity with strength, and that is a meaningful thing for a reader to know. So in this case I would suggest a gendered translation.
Consider also cases which discuss such things as the labor of a breadwinner, or sexual ethics, or kingship. We could render these in a gender neutral way, but that would hide the fact that in that society, these things were thought of in a gendered way. Changing the gender, in that time and place, would quite probably have meant changing the way it was discussed. So I would recommend using gendered language in such cases.
In some cases it is a matter of weight and emphasis, and there is no black and white. Compare, for example, AN 4.122 with AN 5.76. They share a passage about a monk who becomes lustful when he sees a scantily clad woman. In this case, the message can obviously be interpreted from the perspective of both genders, but since it is presented in a gendered context, it’s best to keep the masculine. However, in AN 4.122 this is just one passage in a larger sutta that deals with other, non-gendered, issues. In AN 5.76 it is the main theme. So while it is an easy call to keep the masculine perspective for AN 5.76, AN 4.122 is less clear. Should we make the whole sutta masculine to fit with this one passage? Or should we render masculine in just the relevant passage, and accept the inconsistent voice? Or should we keep the gender-neutral voice throughout, despite the incongruity in this passage? I’ve chosen the last option, but I might change my mind!
Structured lists are not found in the original text, and should be avoided unless there is a substantial advantage in using them. This includes most ordinary lists of things. Remember also that lists take up space, especially relevant for print editions.
The specific advantage lists bring is that they organize information spatially. This is useful in the case of Dhammas presented as a “wheel” (cakka), a common formula in the Aṅguttara and elsewhere. Consider the following:
- Something is A but not B,
- something is B but not A,
- something is both A and B, and
- something is neither A nor B.
Presenting this as a list allows a faster, more intuitive grasp of the structure involved. However this is only really meaningful if the items are fairly short.
If the list is of a set up items that make a meaningful number, as in AN, they should be marked as an ordered list, otherwise unordered.
In the case of a list where all the list items are found in one segment, you can use a simplified markup. Write the sentence as you would normally. Then simply insert a tilde ~ at the start of each list item. SC will take care of the rest when this is converted to HTML. Here is an example segment:
~Here is a cat that is not black, ~here is a cat that is black, ~here is a dog that is not black, and ~here is a dog that is black.
Note that if we remove the tildes, it works perfectly well as a sentence.
This will not work if your list spans multiple segments. In that case, you must use proper HTML list markup. Only use bare markup, avoid "types", "classes", and so on.
In all cases, as with everything you do, take care to observe good practices for writing. Here is a helpful guide to writing lists. These principles may vary in your language.