3 ms·

I've always assumed that somebody must have gone through classifying bugs like this in faulty debate (e.g. based on ancient Greek debating theory, or law, or lo

I've always assumed that somebody must have gone through classifying bugs like this in faulty debate (e.g. based on ancient Greek debating theory, or law, or logic as taught in philosophy?). By using well-defined terms to refer to the bugs, it would be way easier to call them out and mitigate them - as the amount of time wasted in FOSS collaboration with stuff like this is unbelievable. Some egregious examples that spring to mind are: * Trying to disprove a point by finding an edge case where it's not true - even if that edge case is entirely irrelevant (a bit like the complaints in the original post) * Trying to disprove a point by exaggerating it to an extreme and saying that because it doesn't work in the extreme case, it must be false. * Taking a point out of context and proudly proving it to be wrong... when taken out of its wider context. * Making an entirely unrelated point which happens to be true, as a way of trying to prove an earlier point. I guess the second one is a form of hyperbole (but is that the right term)? The third one could be called out as "taken out of context", and fourth is "irrelevant"... but it still feels like life could move on faster if there was a clearer taxonomy for these 'bugs'. (One could even provide the taxonomy as tags/reactions into your communication system of choice - like a more nuanced version of "+5 Insightful" or similar). Edit: perhaps https://en.wikipedia.org/wiki/List_of_fallacies https://en.wikipedia.org/wiki/List_of_fallacies is what i'm looking for, although it feels... cumbersome.

IAmEveryone 7y ago
Yes, of course: "Fallacies". But note there's also something I like to call the "Fallacy Fallacy": People read the wikipedia list and consider them some sort of universal natural law. Case in Point: "appeal to authority". This can indeed be fallacious, within the context where these were formulated, which is probably academic publishing. If someone says they've created an HIV vaccine, it would be very wrong to start using it without clinical trials just because the inventor is some famous researcher. If you try to apply the same to daily life and online debate, you're going to wither and die within a month. Example: you're experiencing chest pain. What do you do? You see a doctor! Do you subject them to a quiz on heart disease, and do you re-run every trial that contributed to your treatment? Of course not -– it would cost trillions and take many lifetimes. Instead, you trust your countries certification scheme, and you trust the scientific establishment. Both are authorities. When you return to your favourite restaurant, you are extrapolating expected future performance based on past experience. in other words: you trust them. Or in even otherer words: you consider them the authority on Fried Chicken and rely on that to guide your decisions. There are only two ways to gain an understanding of the world around you as needed to do anything: direct experience, and what others tell you. The fraction of required knowledge that you can individually verify from first principle is probably somewhere below 1%. For everything else, you need to rely on other's direct experience, by continually fine-tuning individual measures of authority by comparing to direct experience where possible, and integrating with other sources where not.
- Arathorn 7y ago
  right, yup, I guess I was grasping for Fallacies. However, reading through Wikipedia's list of them, I'm finding it fairly hard to find the right ones to apply to my examples - and if I used this nomenclature to try to keep FOSS collaboration productive, I think I'd lose the audience pretty rapidly (or get lost in a meta-argument about fallacies). Hence wondering if there could be a simpler taxonomy for the most common types of bug which pedants use to derail discussion (unintentionally or otherwise).
- INGELRII 7y ago
  Appeal to authority is not not deductively valid and thus it's logical fallacy. On the other hand very little can be learned with deductive reasoning at once. Deductive chains in debate are very short and shortcuts must be used. Authority can be used to weigh or prioritize evidence. If you are not doing serious research into the subject, authority is one heuristic that works extremely well. For example, when two relatively ignorant layman debate in the internet and have limited interest or effort invested to get into the truth. Finding out that something is mainstream view among authorities in the field should end the argument for the time being for the lack of more information. Unfortunately the discussion usually ends with questioning that particular field, then the progress of the society as a whole. Scientists have been wrong before so .. they can also be corrupt ...
- travisjungroth 7y ago
  Logic means two things: “Reasoned thinking” and “a strict set of rules for proving things” When someone wants to disagree, it’s easy to just sub out the definitions and say something isn’t logical because it doesn’t follow formal methods. Believing what someone says because they’re well-respected about a topic isn’t strict, but it’s really useful.
matwood 7y ago
Most of these issues (yours, the OP, and others here in the comments) are all covered under the Principal of Charity[1]. Unfortunately, this is the internet and people would rather feel like they won something (unsure exactly what), rather than have a real discussion where theirs or someone else's mind can be changed. While not perfect, I do find HN to be a better place for discussion for most topics. [1] https://en.wikipedia.org/wiki/Principle_of_charity https://en.wikipedia.org/wiki/Principle_of_charity

IAmEveryone 7y ago

Yes, of course: "Fallacies". But note there's also something I like to call the "Fallacy Fallacy": People read the wikipedia list and consider them some sort of universal natural law. Case in Point: "appeal to authority". This can indeed be fallacious, within the context where these were formulated, which is probably academic publishing. If someone says they've created an HIV vaccine, it would be very wrong to start using it without clinical trials just because the inventor is some famous researcher. If you try to apply the same to daily life and online debate, you're going to wither and die within a month. Example: you're experiencing chest pain. What do you do? You see a doctor! Do you subject them to a quiz on heart disease, and do you re-run every trial that contributed to your treatment? Of course not -– it would cost trillions and take many lifetimes. Instead, you trust your countries certification scheme, and you trust the scientific establishment. Both are authorities. When you return to your favourite restaurant, you are extrapolating expected future performance based on past experience. in other words: you trust them. Or in even otherer words: you consider them the authority on Fried Chicken and rely on that to guide your decisions. There are only two ways to gain an understanding of the world around you as needed to do anything: direct experience, and what others tell you. The fraction of required knowledge that you can individually verify from first principle is probably somewhere below 1%. For everything else, you need to rely on other's direct experience, by continually fine-tuning individual measures of authority by comparing to direct experience where possible, and integrating with other sources where not.

Arathorn 7y ago
right, yup, I guess I was grasping for Fallacies. However, reading through Wikipedia's list of them, I'm finding it fairly hard to find the right ones to apply to my examples - and if I used this nomenclature to try to keep FOSS collaboration productive, I think I'd lose the audience pretty rapidly (or get lost in a meta-argument about fallacies). Hence wondering if there could be a simpler taxonomy for the most common types of bug which pedants use to derail discussion (unintentionally or otherwise).
INGELRII 7y ago
Appeal to authority is not not deductively valid and thus it's logical fallacy. On the other hand very little can be learned with deductive reasoning at once. Deductive chains in debate are very short and shortcuts must be used. Authority can be used to weigh or prioritize evidence. If you are not doing serious research into the subject, authority is one heuristic that works extremely well. For example, when two relatively ignorant layman debate in the internet and have limited interest or effort invested to get into the truth. Finding out that something is mainstream view among authorities in the field should end the argument for the time being for the lack of more information. Unfortunately the discussion usually ends with questioning that particular field, then the progress of the society as a whole. Scientists have been wrong before so .. they can also be corrupt ...
travisjungroth 7y ago
Logic means two things: “Reasoned thinking” and “a strict set of rules for proving things” When someone wants to disagree, it’s easy to just sub out the definitions and say something isn’t logical because it doesn’t follow formal methods. Believing what someone says because they’re well-respected about a topic isn’t strict, but it’s really useful.

matwood 7y ago

Most of these issues (yours, the OP, and others here in the comments) are all covered under the Principal of Charity[1]. Unfortunately, this is the internet and people would rather feel like they won something (unsure exactly what), rather than have a real discussion where theirs or someone else's mind can be changed. While not perfect, I do find HN to be a better place for discussion for most topics. [1] https://en.wikipedia.org/wiki/Principle_of_charity https://en.wikipedia.org/wiki/Principle_of_charity