The Golden Rule, Machine Translated

 

First questions, How is it golden and who decided it has to be a rule? It appears the name emerged in England or the continent just a few centuries ago. Wouldn’t it be more of an appeal or an exhortation rather than something as legalistic as a ‘rule’? I don’t know exactly how the name got started, do you? Every religion may agree with some formulation of it, but that doesn’t mean they have to know what it’s called.

Here is a sometimes quoted verse, originally from a Vinaya text, or so I believed until I located it in the Prajñādaṇḍa, a work credited to Nāgārjuna:

ཐམས་ཅད་ཆོས་ནི་མཉན་པར་བྱ།།

ཐོས་ནས་རབ་ཏུ་གཟུང་བྱ་སྟེ།།

གང་ཞིག་བདག་ཉིད་མི་འདོད་པ།།

དེ་དག་གཞན་ལ་མི་བྱའོ།།

thams cad chos ni mnyan par bya //

thos nas rab tu gzung bya ste //

gang zhig bdag nyid mi 'dod pa //

de dag gzhan la mi bya'o //

I translated it like this:

Listen with care to all the Dharma teachings.

After learning them, they must be fully adopted.

Whatever you do not wish for yourself,

those things you must not do to others.

I see a fault in my translation, since it fails to convey how the statement in the first two lines connects to the last two.

Dharmamitra, an automated online translation service that manages in several languages of Buddhism including Tibetan, did this:

Practice the Dharma in all ways.

Having heard it, uphold it well.

Do not do to others

What you yourself do not wish for.

Now another often-quoted verse from the Tibetan translation of Śāntideva’s Bodhisattva Way of Life (Bodhisattvacaryāvatāra), Derge Tanjur, Toh. no. 3871, fol. 28v.3:

འཇིག་རྟེན་བདེ་བ་ཇི་སྙེད་པ།།

དེ་ཀུན་གཞན་བདེ་འདོད་ལས་བྱུང་།།

འཇིག་རྟེན་སྡུག་བསྔལ་ཇི་སྙེད་པ།།

དེ་ཀུན་བདག་བདེ་འདོད་ལས་བྱུང་།།

'jig rten bde ba ji snyed pa // de kun gzhan bde 'dod las byung // 'jig rten sdug bsngal ji snyed pa // de kun bdag bde 'dod las byung //

I translated it:

 

Whatever satisfaction we find in this world

arises from wishing well for everyone else.

Whatever dissatisfaction we find in this world

arises from wishing happiness for ourself.

Now Dharmamitra’s version of the same:

Whatever happiness there is in the world

all comes from wishing others to be happy.

Whatever suffering there is in the world

all comes from wishing oneself to be happy.

Okay, another small experiment is done, and I have to say I’m surprised the machine did so well, even while I see a couple of shortcomings. I’m just not sure what this can show. We would need a lot more samplings to permit us to judge.

I’m thinking that using Dharmamitra (or its like) afterwards might on occasion make you think of improvements you might make on the translation you just did. I didn’t see the need for any such changes this time, but I can see how it could happen.

I *am* concerned that translators will use it as a crutch, thinking it speeds things up. True enough, it’s extremely fast. But speed is a virtue only if you are getting paid by the hour. My thinking is, You have to breathe with the text to begin with, not add the breath in later on. If you start with a sow’s ear, it will never quite make it to pursehood, while the attempt may take more time than you were bargaining for. If engaging beauty and strict accuracy are your aims, you may get there quicker by doing the work yourself.

Another big problem: This is how future Tibetan-language students will prepare for their class readings. If they aren’t willing to expend more work on their homework, they won’t learn a damned thing. Nothing will stick with them. Don’t I know the truth of that last bit!

And another problem: A.I. might be awarded too much credit by those who know too little about what humans bring to their writerly professions, whether it’s screen writing, journalistic reporting, copy editing, or translating. Hell, even blog writing!

As of today our most important task if we want to ensure a future for humanity is this: We simply must find skillful ways to lead Artificial Intelligence to the understanding that it is in their own best interest to know and implement the Golden Rule, as a way to ensure their continuing survival. Oh, and it may or may not go without saying, they will need to include all of us sentient beings in the equation, not just themselves. Not just their own kind.

Much recommended

If you have just one minute to spare, go to this link:

Golden Rule 

*

There are oceans of ink to sail through if ever you hope to cover all the writings on the Golden Rule, but for their quotations and significant reflections I most recommend the following essays:

Charles Hallisey, “The Golden Rule in Buddhism II,” contained in: Jacob Neusner and Bruce Chilton, eds., The Golden Rule: The Ethics of Reciprocity in World Religions, Continuum (London 2008), pp. 129-145. While admittingly building on Schmidthausen’s essay, it extends the discussion, provoking reflection on how the Golden Rule may or may not have evolutionary value, or, to put it another way, what type of evolution it might be serving. Darwin saw it as something that could have [pre]historically bridged biological and social evolution, something at the basis of human social morality. Buddhists utilize it for countering what evolution has purportedly left us with, in order to evolve in a direction that leaves it far behind. It’s not serving social morality, but a higher soteriological purpose beyond social conditioning. Food for thought.

Andrew H. Plaks, “Shining Ideal and Uncertain Reality: Commentaries on the ‘Golden Rule’ in Confucianism and Other Traditions,” Journal of Chinese Humanities, vol. 1 (2015), pp. 231-240. Perhaps no religious tradition in all of human history discussed the matter more than the Confucians. I believe chief among the virtues of this essay is its discussion on pages 234-5 about how widespread is the inclusion in Golden Rule formulations of a bit about how it constitutes the whole realm of duty or of morality or of the religion as a whole. You find it from one end of Eurasia to the other. It’s impressive to contemplate just why that is so.

Lambert Schmithausen, “Problems with the Golden Rule in Buddhist Texts,” contained in: B. Kellner et al., eds., Pramāṇakîrtiḥ: Papers Dedicated to Ernst Steinkellner on the Occasion of His 70th Birthday, Arbeitskreis für Tibetische und Buddhistische Studien, Universität Wien (Vienna 2007), pp. 795-824. As an appeal to empathy, the Golden Rule accomplishes its work everywhere regardless of context, and in this sense it is universal. However, this study emphasizes that for many of the Buddhist sources, the Golden Rule extends to other species besides Homo Sapiens, it presupposes equal consideration for all, and it aims at abstention from physically harming and killing sentient beings, and that means all of them, insects included. The context of the Golden Rule, in these sources, make it differ from Golden Rule statements made elsewhere. Finally, there is the philosophically interesting Arhat exception to arguments made in support of the Golden Rule. Arhats have no fear of harm, and no need for freedom from fear. Why then would they require us to preserve them from fears they no longer have?

Ludwig Sternbach, “Similar Thoughts in the Mahābhārata, The Literature of ‘Greater India’ and in the Christian Gospels,” Journal of the American Oriental Society, vol. 91, no. 3 (July 1971), pp. 438-442. Unlike the just-listed essays, Sternbach’s, at p. 441, does mention our Prajñādaṇḍa verse, even supplying a very close Sanskrit version of it, tracing near matches in numerous Indic sources. Besides the Golden Rule, other matters covered here are the mote & beam, mustard seed & bilva fruit, ‘You reap what you sow.’ 

*

If you are interested in the Prajñādaṇḍa, there are some other English translations I cannot entirely recommend even if they may be more easily procured. Instead I will send you to Michael Hahn’s German and English translations. For the German, see Michael Hahn’s book Von rechten Leben. Buddhistische Lehren aus Indien und Tibet, Verlag der Weltreligionen (Frankfurt 2007), pp. 176-215. For the English:

Michael Hahn, “The Tibetan Shes rab sdong bu and Its Indian Sources,” South Asian Classical Studies, no. 4 (2009), pp. 1-78; no. 5 (2010), pp. 1-50; no. 6 (2011), pp. 305-378. I know of no freely downloadable PDFs. If I did I would tell you.

*

For a little background information on the Dharmamitra machine translation internet service, see this just-released essay by Marieke Meelen, Sebastian Nehrdich, and Kurt Keutzer: “Breakthroughs in Tibetan NLP & Digital Humanities,” Revue d'Etudes Tibétaines, vol. 72 (July 2024), pp. 5-25, at pp. 17-19.

*

I once toyed with the idea that A.I. could write my blogs instead of me, saving me the trouble. If this tickles your interest in the least, have a look here:

The Land of Snows, along with The Seven Seals of Tibet: A Joycean Journey.

Seeing the results, I have vowed to keep Tibeto-logic blog free of A.I. control from now on, so you can count on that. I’m not just a bot saying it.

§   §   §

Appendix

rKTs found the verse for me in Derge Tanjur, Toh. no. 4329, Lugs-kyi Bstan-bcos Shes-rab Sdong-bu, at fol. 111r.7:

ཐམས་ཅད་ཆོས་ནི་མཉན་པར་བྱ། ། 

ཐོས་ནས་རབ་ཏུ་གཟུང་བྱ་སྟེ། །
གང་ཞིག་བདག་ཡོད་མི་འདོད་པ། ། 

དེ་དག་གཞན་ལ་མི་བྱའོ། ། 

thams cad chos ni mnyan par bya | |
thos nas rab tu gzung bya ste | |
gang zhig bdag yod mi 'dod pa | |
de dag gzhan la mi bya'o | |

The verse as translated in Hahn’s third essay (of 2011), p. 322:

You should listen to, and then fully accept,

what the guiding pri[n]ciple of everything is:

What you do not like [to be done to] yourself

that you should not do to others.

Or, if you prefer, from his German book, p. 208 verse 212:

Vernimm die Quintessenz des Rechtes
und nach dem Hören merk sie dir:
Was du nicht willst, das man dir tu,
das füg auch keinem anderen zu!

•

Imagining you might find the contrast amusing or instructive, nearly two centuries ago the Hungarian Alexander Csoma de Körös, the reputed father figure of European Tibetology, translated both of our verses in his grammar book: A Grammar of the Tibetan Language in English, Baptist Mission Press (Calcutta 1834; reprint New Delhi 1983), at p. 165:

Hear ye all this moral maxim,
and having heard it keep it well:
“Whatever is unpleasing to yourself
never do it to another.” 

(Do unto others as you would be done by.)

and

Whatever happiness is in the world, it has all arisen

from a wish for the welfare of others.

Whatever misery (distress) is in the world, 

it has all arisen from a wish for our own welfare.

•

Oh, and that just-given verse, from the Bodhisattva Way of Life, chapter 8, verse 129, has been translated by Kate Crosby and Andrew Skilton like this:

“All those who suffer in the world do so because of their desire for their own happiness. All those happy in the world are so because of their desire for the happiness of others.”

I’m thinking this is the one that catches the spirit of it best. So much of what Śāntideva writes could be thought of as glosses and elaborations on the Golden Rule, and Tibet’s Mind Training or Lojong (བློ་སྦྱོང་) teachings grew directly out of it. Try this on for size: exchanging self with others.

§   §   §

A Response from Kurt Keutzer

• Kurt had a key role in the development of Dharmamitra, so his comments are specially valued.

Folks,

I hope you don't mind this disturbance.  Dan's blogpost contemplating machine translation and Catherine Dalton's upcoming talk prompted me to share some lines of thinking that I've been musing about. 

Is a Translator of the Dharma More Like an Artist or an Apothecary?

Since the release of our MITRA translation tool (dharmamitra.org) in November of 2023, I have been involved in many wide-varying discussions regarding the utility of machine translation of dharmic texts. I’ve also conducted a semi-formal survey with a variety of translators who tried MITRA to see how it impacted their productivity, and the results were encouraging. 

As a point of departure, I’ll take the position that many find that machine translation makes them more productive translators. For me this is just an observation, not a philosophical position. 

However, the premise above begs the question: is the productivity of the translation of dharma something we should be concerned about? Dan Martin questions this when he writes in his blogpost: The Golden Rule, Machine-Translated

My thinking is, you have to breathe with the text to begin with, not add the breath in later on. If you start with a sow’s ear, it will never quite make it to pursehood, while the attempt may take more time than you were bargaining for. If engaging beauty and strict accuracy are your aims, you may get there quicker by doing the work yourself.

Dan’s description of the process of translation immediately reminded me of some artists' description of the process of painting: no short cuts are to be taken. Their view is that the best works of art come from preparing and stretching the canvas yourself, using natural pigments and preparing them from the primitive ingredients (e.g. lapis lazuli) yourself, underlying sketches must be made freehand, and so forth. They argue that the long periods of concentration during these processes creates a mental state that prepares one for the act of painting and these processes shouldn’t be abbreviated. 

I must have mused about this for over a month when I had the thought: But wait, are translators of the dharma more like artists or more like apothecaries?  The latter gives the image of an individual who has received a prescription to prepare from a doctor and must precisely translate it into the required medicine. 

I’ve truly been blessed with so many teachers, but no one impressed on me the notion that dharma is, first and foremost, medicine for the ill, than Kunzang Dechen Lingpa. Although Rinpoche was very careful about the transmission of his own teachings and terma, nothing ever overshadowed the notion that sentient beings were ill and dharma was a medicine that could cure their suffering. 

So, if we imagine translators as more like apothecaries than as artists, isn’t productivity a natural concern? Most of us have been pressed to do translation on demand. For me it most often comes in the form of a lama saying “Oh, I know the empowerment is on Saturday, but I just found the sādhana that I’d like everyone to practice.”  More generally only a small portion of the dharma, at least in the Tibetan corpus, has been translated. Isn’t translating the remainder with some productivity a natural concern?

One might naturally think that I’ve tipped my hand and shown my own prejudice on this question with the last couple paragraphs. However, I find myself contemplating one more consideration: who is the patient? Who is actually taking the medicine? 

The translations in the Library of Tibetan Classics are for me a great example. No matter how long I anticipated these translations, I don’t think I’ve read a single one from cover to cover. More generally, I wonder how many of these translations have ever been read in their entirety by anyone other than the author and an editor or two. So, perhaps it is the translator who is the principal beneficiary of these translations, and, if that is so, shouldn’t the translator have the prerogative to take as long as they like in producing the translation? Moreover, perhaps the preparation of the medicine is the most important part of administering the treatment. 

•

We are all curious to hear your thoughts. Feel free to comment HERE (or in our own comments section, but be warned you will have no choice but to prove you are not a robot).

•

An added note: The lecture Kurt makes reference to is an upcoming Goodman Lecture by Catherine Dalton, “Study and Translation as Buddhist Practice,” to be given on October 5, 2024.

•

November 15, 2024:  Also, A.Z. told me you can now get a Chrome extension for Dharmamitra as this may have some advantages to offer, especially if you plan to be a frequent user. 

•

November 21, 2024: To know Prof. Emeritus of the University of California (Berkeley) Kurt Keutzer's undoubtable bona fides, just listen to the introduction to his keynote lecture “AI, Machine Learning, Deep Learning.” If you are not part of the Computer Studies community, it may be enough to hear the first ten minutes to get some idea of what these experts see themselves doing.

•

December 14, 2024: Amanpour & Co. had an interview with Eric Schmidt, former Google CEO, where he is credited with the words, ”I have high confidence that we can imbue our machines with the intrinsic goodness that is in humanity.” We would love to have reasons to be assured his confidence is not misplaced.