{"id":875,"date":"2026-10-05T23:34:11","date_gmt":"2026-10-05T23:34:11","guid":{"rendered":"https:\/\/morganredfield.com\/blog\/?p=875"},"modified":"2026-10-05T23:36:40","modified_gmt":"2026-10-05T23:36:40","slug":"representations-and-core-of-ethics","status":"publish","type":"post","link":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/","title":{"rendered":"Representations of ethics; core of ethics"},"content":{"rendered":"\n

Consequentialism, virtue ethics, and deontology are often presented as being different ethics. I’ve been in a few arguments about whether consequentialism is better than virtue ethics, and most people I know prefer either to the rigid rules of deontology. After reading various philosophy blogs online, I ended up spending a long time thinking about these and trying to really feel how consequentialism was better from the inside.<\/p>\n\n\n\n

Ultimately, the position I came to was that these three types of ethics were just different representations of “the good”. They have different pros and cons as representations, but any of them could be replaced by a specific instantiation of the others that could approximate it to an arbitrary degree of accuracy.<\/p>\n\n\n\n

It’s possible to convert from one representation to the other in a few ways. The most trivial is just to wrap one representation with another, such as “the rules say be a virtuous person”. A more nuanced version of that conversion would actually select rules such that, by following them, you incidentally embody virtue. The following table shows a high level concept for what this looks like.<\/p>\n\n\n\n

From \\ To<\/th>deontology<\/th>virtue<\/th>consequentialism<\/th><\/tr><\/thead>
deontology<\/strong><\/td>–<\/td>choose virtues such that embodying them leads to compliance with the given ruleset<\/td>utility function is defined such that worlds where the agent happens to follow the ruleset are highly valued<\/td><\/tr>
virtue<\/strong><\/td>choose rules such that following them leads to embodying the set of virtues<\/td>–<\/td>utility function is defined such that world states in which agent has virtue are highly valued<\/td><\/tr>
consequentialism<\/strong><\/td>choose rules such that following them leads to the maximization of utility<\/td>choose virtues (such as charity) such that embodying them maximizes utility<\/td>–<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

Pros and Cons<\/h2>\n\n\n\n

I originally learned about consequentialism as a foundation for ethics from people doing artificial intelligence work. When doing model based planning for a Markov Decision Process, using consequentialism is quite straightforward. Observe the world, check if you like the world, predict the effect of all your actions, choose the action that you predict will lead to the best world. Repeat.<\/p>\n\n\n\n

For large worlds this becomes computationally intractable. For partially observable worlds you need to get more complex. In any case, you don’t have to program in any kind of rules or virtues. You just say what “gets you points” and let a reinforcement learning optimization algorithm figure out how to act.<\/p>\n\n\n\n

The problem with this is that our world is both large and only partially observable. Not only that, there is no easily observable “score” that humans can use to tell if they’re doing good or bad. We can rely on feelings to some extent, but those are notoriously fickle and scale insensitive.<\/p>\n\n\n\n

With a deontological ethics, humans can have rules set forth in an intuitive way. Don’t murder, don’t steal. These are clear to us, even if they would be hard to fully describe in a utility function that was only based on observable world states (Are executions murder? What about war?).<\/p>\n\n\n\n

The problem with a set of rules is that it’s limited. Go to a new context and it can be hard to tell what applies. This is where virtue ethics comes in. We don’t need a ruleset, we just try to be good people. We don’t need to observe huge portions of the world and do difficult prediction work, we just try to be good people. But this can lead us to be self-centered, and to focus too much on being good people and not enough on doing good works.<\/p>\n\n\n\n

Information and Compute Advantages<\/h2>\n\n\n\n

I’ve heard it argued that these three frameworks are good for different informational environments. If you don’t know very much about your world, simple rules are best (deontology). If you know a bit more, but have trouble predicting complex systems, virtue ethics is best. If you can observe and predict the world well, consequentialism is best.<\/p>\n\n\n\n

The implication here is that as we get smarter and better at understanding the world, we should move more towards consequentialism. I agree with this in some ways, but I also think it avoids much of the complexity of ethics.<\/p>\n\n\n\n

The standard rules of deontology only seem simple to us because they’re framed in human terms. Giving a robot the ten commandments is actually very difficult (easier now with LLMs, but they’ve read the entire internet in order to learn how to think as humans think). Deontology only seems simple to us because the rules are intentionally framed in a humanly understandable way, and even then there are ancient traditions examining what the rules “really mean”.<\/p>\n\n\n\n

In fact, one could argue that as robots and artificial systems have gotten more intelligent or learned more, they have moved away from a consequentialism (especially for things like Constitutional AI<\/a>). Representing an accurate reward function was more complicated than using a list of principles to follow.<\/p>\n\n\n\n

I no longer think there’s an inherent information advantage to one of these ethical theories. Instead, the simplest representation depends on the agent, the world that they’re living in, and the ethics that they’re trying to think about.<\/p>\n\n\n\n

Natural Basis<\/h2>\n\n\n\n

The difference between the representations goes deeper than the information needed to perform them well. Each of the representations actually does privilege a specific ethical outcome. That’s part of why I spent so long trying to decide if I was a virtue ethicist or a consequentialist. They can<\/em> describe the same ethics, but certain ethical theories are easier to think about from a specific framework.<\/p>\n\n\n\n

Look again at the table from the top of this post. It shows how to convert from one ethical framework to another, but the conversions seem a bit forced. Converting from consequentialism to deontology involves rules that say that you should do things that happen to maximize utility. That looks something like “give 10% of your income to the most impactful charity.” By making the rule more complex, you could more closely approximate consequentialism, but it’ll never be as simple as the utility maximizing framework is.<\/p>\n\n\n\n

Another example is that virtue ethics privileges ethical theories in which there is a moral agent. It doesn’t just matter what that moral agent does, it matters why. Approximating this with consequentialism requires a much more complex description.<\/p>\n\n\n\n

I think this is why we can get so caught up in questions about which is correct. It’s much easier to think about certain ethical outcomes from a given representation. Despite that, you can still formulate and think about very different outcomes from the same representation. That makes arguments about them inherently confusing unless participants are very clear about what they’re talking about.<\/p>\n\n\n\n

I spent too much time thinking about consequentialism vs virtue ethics. I would have been better served by thinking about what my utility function would be if I were consequentialist. Or what virtues I would want to embody. Or even what rules should guide me as a deontology.<\/p>\n\n\n\n

What is ethics for?<\/h2>\n\n\n\n

I sometimes see people claiming that ethics is just game theory. To me this misses a huge portion of what ethics is about. If rationality is the art of winning, then ethics is the art of doing what’s right. Even if you were the only agent in the universe, there would still be a right thing to do. An emphasis on game theory makes sense because it impacts our day to day experience of ethical decisions so much, but it risks eliding the importance of just what we are helping others for<\/em>.<\/p>\n\n\n\n

There’s a huge focus right now on helping others. Sometimes that manifests as the idea of ruining a suit to save a life, other times it’s about cooperating with similar minds in prisoner’s dilemmas. In these cases and more, the focus is on what we owe others and on how much we should (or must) help them.<\/p>\n\n\n\n

This is all very good. It’s good to think about how to be charitable, to help others. It’s good, but I don’t think it works well as a starting point. If you’re helping someone else, what<\/em> are you helping them to do?<\/p>\n\n\n\n

Consequentialists need to know what consequences they value. Virtue ethicists need to know what virtues they are trying to embody. Deontologists need to know what rules they’re following. Regardless of which representation you pick to think about ethics, you need to know what you think is good.<\/p>\n\n\n\n

Lots of conversations these days seem to take this for granted, but I think it’s actually pretty difficult to know what is best in life.<\/p>\n\n\n\n

The good life<\/h2>\n\n\n\n

One thing I missed during my rationalist years learning about consequentialism and game theory was the search for what was good in my own life.<\/p>\n\n\n\n

I’ve gotten a lot more traction on ethics by thinking about what a good life is. It’s easy to say that you should save someone’s life. Living is better than dying in most circumstances. It’s harder to ask how to help someone live well. What would be a good life for them? How could you help them with that? And hey, what would be a good life for you?<\/p>\n\n\n\n

It’s become easier for me to be generally charitable as I’ve become more satisfied with my own life. Easier for me to feel a lot of care for others. Easier for me to give folks good life advice, instead of just giving them money. Easier for me to engage with my community in a courageous ethical way, in addition to quietly sending money to the desperate in other countries.<\/p>\n\n\n\n

I think this idea of “a good life” is the core to ethics. I have some answers now for what I think the good life is for me, and some hints about what more answers might look like. This keeps me grounded when I think about ethics, from whatever representation.<\/p>\n\n\n\n

I worry that the current focus on ethics as “what we owe others” skips over some of the very important questions in life.<\/p>\n\n\n\n

We need to discover what is right for our own lives, not just calculate the optimal amount to help others. Using all of the representations of ethics can help with that.<\/p>\n","protected":false},"excerpt":{"rendered":"

Consequentialism, virtue ethics, and deontology are often presented as being different ethics. I’ve been in a few arguments about whether consequentialism is better than virtue ethics, and most people I know prefer either to the rigid rules of deontology. After reading various philosophy blogs online, I ended up spending a long time thinking about these … Continue reading “Representations of ethics; core of ethics”<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-875","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"yoast_head":"\nRepresentations of ethics; core of ethics - Theory and Practice<\/title>\n<meta name=\"description\" content=\"Consequentialism, virtue ethics, and deontology form separate views onto the same core of a good life.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Representations of ethics; core of ethics - Theory and Practice\" \/>\n<meta property=\"og:description\" content=\"Consequentialism, virtue ethics, and deontology form separate views onto the same core of a good life.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/\" \/>\n<meta property=\"og:site_name\" content=\"Theory and Practice\" \/>\n<meta property=\"article:published_time\" content=\"2026-10-05T23:34:11+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-10-05T23:36:40+00:00\" \/>\n<meta name=\"author\" content=\"morgan\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"morgan\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"8 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/\"},\"author\":{\"name\":\"morgan\",\"@id\":\"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd\"},\"headline\":\"Representations of ethics; core of ethics\",\"datePublished\":\"2026-10-05T23:34:11+00:00\",\"dateModified\":\"2026-10-05T23:36:40+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/\"},\"wordCount\":1728,\"publisher\":{\"@id\":\"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd\"},\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/\",\"url\":\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/\",\"name\":\"Representations of ethics; core of ethics - Theory and Practice\",\"isPartOf\":{\"@id\":\"https:\/\/morganredfield.com\/blog\/#website\"},\"datePublished\":\"2026-10-05T23:34:11+00:00\",\"dateModified\":\"2026-10-05T23:36:40+00:00\",\"description\":\"Consequentialism, virtue ethics, and deontology form separate views onto the same core of a good life.\",\"breadcrumb\":{\"@id\":\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/morganredfield.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Representations of ethics; core of ethics\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/morganredfield.com\/blog\/#website\",\"url\":\"https:\/\/morganredfield.com\/blog\/\",\"name\":\"Theory and Practice\",\"description\":\"Exploring life through science and art.\",\"publisher\":{\"@id\":\"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/morganredfield.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd\",\"name\":\"morgan\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/0da7fa0a0a07a8497d51de4300111d978b649e254fc1afa26dc67f3b83d24c38?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/0da7fa0a0a07a8497d51de4300111d978b649e254fc1afa26dc67f3b83d24c38?s=96&d=mm&r=g\",\"caption\":\"morgan\"},\"logo\":{\"@id\":\"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/image\/\"},\"url\":\"https:\/\/morganredfield.com\/blog\/author\/morgan\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Representations of ethics; core of ethics - Theory and Practice","description":"Consequentialism, virtue ethics, and deontology form separate views onto the same core of a good life.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/","og_locale":"en_US","og_type":"article","og_title":"Representations of ethics; core of ethics - Theory and Practice","og_description":"Consequentialism, virtue ethics, and deontology form separate views onto the same core of a good life.","og_url":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/","og_site_name":"Theory and Practice","article_published_time":"2026-10-05T23:34:11+00:00","article_modified_time":"2026-10-05T23:36:40+00:00","author":"morgan","twitter_card":"summary_large_image","twitter_misc":{"Written by":"morgan","Est. reading time":"8 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/#article","isPartOf":{"@id":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/"},"author":{"name":"morgan","@id":"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd"},"headline":"Representations of ethics; core of ethics","datePublished":"2026-10-05T23:34:11+00:00","dateModified":"2026-10-05T23:36:40+00:00","mainEntityOfPage":{"@id":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/"},"wordCount":1728,"publisher":{"@id":"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd"},"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/","url":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/","name":"Representations of ethics; core of ethics - Theory and Practice","isPartOf":{"@id":"https:\/\/morganredfield.com\/blog\/#website"},"datePublished":"2026-10-05T23:34:11+00:00","dateModified":"2026-10-05T23:36:40+00:00","description":"Consequentialism, virtue ethics, and deontology form separate views onto the same core of a good life.","breadcrumb":{"@id":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/morganredfield.com\/blog\/2026\/10\/representations-and-core-of-ethics\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/morganredfield.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Representations of ethics; core of ethics"}]},{"@type":"WebSite","@id":"https:\/\/morganredfield.com\/blog\/#website","url":"https:\/\/morganredfield.com\/blog\/","name":"Theory and Practice","description":"Exploring life through science and art.","publisher":{"@id":"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/morganredfield.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":["Person","Organization"],"@id":"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/15fce334735dd60ce7bb5712c44a62dd","name":"morgan","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/0da7fa0a0a07a8497d51de4300111d978b649e254fc1afa26dc67f3b83d24c38?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/0da7fa0a0a07a8497d51de4300111d978b649e254fc1afa26dc67f3b83d24c38?s=96&d=mm&r=g","caption":"morgan"},"logo":{"@id":"https:\/\/morganredfield.com\/blog\/#\/schema\/person\/image\/"},"url":"https:\/\/morganredfield.com\/blog\/author\/morgan\/"}]}},"_links":{"self":[{"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/posts\/875","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/comments?post=875"}],"version-history":[{"count":2,"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/posts\/875\/revisions"}],"predecessor-version":[{"id":878,"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/posts\/875\/revisions\/878"}],"wp:attachment":[{"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/media?parent=875"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/categories?post=875"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/morganredfield.com\/blog\/wp-json\/wp\/v2\/tags?post=875"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}