Log in
Register
Search
Search titles only
By:
Search titles only
By:
Forums
New posts
Forum list
Search forums
Leaderboards
Games
Our Blog
Blogs
New entries
New comments
Blog list
Search blogs
Credits
Transactions
Shop
Blessings: ✟0.00
Tickets
Open new ticket
Watched
Donate
Log in
Register
Search
Search titles only
By:
Search titles only
By:
More options
Toggle width
Share this page
Share this page
Share
Reddit
Pinterest
Tumblr
WhatsApp
Email
Share
Link
Menu
Install the app
Install
Forums
Discussion and Debate
Discussion and Debate
Physical & Life Sciences
Intent
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Reply to thread
Message
<blockquote data-quote="FrumiousBandersnatch" data-source="post: 73541878" data-attributes="member: 241055"><p>This is the perennial problem with our categorization of the natural world, a generalisation of the sorites <a href="https://en.wikipedia.org/wiki/Sorites_paradox" target="_blank">paradox</a> - how can one distinguish separate categories in what are, for all intents and purposes, continua? The answer seems to be that we generally take a utilitarian approach and centre categories around the most common or striking values or types in our everyday experience, using whatever suitable delineating boundaries we can establish, often fuzzy. For example, we do this with 'age', and the electromagnetic spectrum, which are fairly simple one-dimensional categories; but we also do it with categories that are fuzzy in many dimensions, like 'life', 'alive', 'species', 'mind', etc.</p><p></p><p>In my experience, intent is generally reserved for organisms that direct their behaviour in ways recognisably similar to our own, i.e. that have minds, some level of consciousness, and that 'think' (i.e. sophisticated information processing), using a specialised organ (brain) that can map the world and produce an internal model to predict and select future behaviour strategies based on the stored results of previous behaviour (learning, knowledge). Basing behaviour on a representation or abstraction of the world, on dynamic modelling, gives it a level of indirection, or '<a href="https://en.wikipedia.org/wiki/Aboutness" target="_blank">aboutness</a>' that is associated with intentionality.</p><p></p><p>I'm not saying that this is how people decide whether to call an action intentional, but that it seems to me that when they use the concept literally, they're usually implicitly distinguishing that category. It seems to me that, to keep its semantic utility, the literal use of intent should be restricted to that category, and when the concept is used to describe behaviour outside it, it is anthropomorphic projection using the <a href="https://en.wikipedia.org/wiki/Intentional_stance" target="_blank">intentional stance</a>, i.e. viewing behaviour in terms of mental properties that are not actually present, but that, as cognitively familiar terms, help us construct a meaningful or interesting narrative for it.</p><p></p><p>But that is itself an example of anthropomorphic projection, the intentional stance writ large. An amoeba doesn't have the capacity for thought - a cognitive sense of self, or capacity for planning and forethought; if that level of cognitive capability was possible without a brain, we wouldn't need or have brains.</p><p></p><p>This is why it's I think it's important to recognise when the intentional stance is used; e.g. a raindrop on the window isn't really <em>contemplating</em>, then <em>deciding,</em> then <em>choosing</em> the path of least resistance down the window, a thermostat doesn't really think "Is it warm enough for me to switch off yet?", then make a <em>decision </em>to switch when it <em>thinks </em>the temperature is just about right.</p><p></p><p>Infants and children seem to be primed to interpret events in the world in terms of anthropomorphic narratives (i.e projected sentient agency), which is probably a by-product of innate predisposition to <a href="https://theness.com/neurologicablog/index.php/hyperactive-agency-detection/" target="_blank">Hyperactive Agency Detection</a>, and their developing <a href="https://en.wikipedia.org/wiki/Theory_of_mind" target="_blank">theory of mind</a>. This tendency <a href="https://blogs.scientificamerican.com/thoughtful-animal/animating-anthropomorphism-giving-minds-to-geometric-shapes-video/" target="_blank">persists into adulthood</a>, by which time the distinction between the reality and the anthropomorphic narrative is usually well understood; nevertheless, the anthropomorphic narrative is often the preferred way to interpret activity, and can be a source of entertainment or amusement (as in cartoons and movies of animals behaving like humans, and the predilection for dressing up pets).</p><p></p><p>'Chaos' occurs when the output of a system is sensitively dependent on its initial conditions; it's an exponential divergence in output or outcome for arbitrarily small differences in input or starting conditions, but it is deterministic.</p><p></p><p>Given that amoebae use external chemical concentration gradients to detect food particles and organise their movement, and that their internals are fairly dynamic, it wouldn't surprise me if putting amoebae into a uniform environment, i.e. even lighting conditions, without chemical gradients, etc., would produce pseudo-random or chaotic movement, as any movement would result from small unpredictable variations in their internal conditions.</p><p></p><p>Chaotic behaviour seems to me a rather <em>poor</em> indicator of intent - all kinds of simple natural phenomena display chaotic behaviour, from <a href="https://en.wikipedia.org/wiki/Brownian_motion" target="_blank">Brownian motion</a> to the path of lightning strikes. It seems more likely that we should expect to see a relatively <em>low</em> incidence of chaotic behaviour in living things, as goal-oriented behaviours are more likely to tend towards common paths or solutions to particular goals (although an injection of chaotic behaviour is sometimes used to make life difficult for predators).</p></blockquote><p></p>
[QUOTE="FrumiousBandersnatch, post: 73541878, member: 241055"] This is the perennial problem with our categorization of the natural world, a generalisation of the sorites [URL='https://en.wikipedia.org/wiki/Sorites_paradox']paradox[/URL] - how can one distinguish separate categories in what are, for all intents and purposes, continua? The answer seems to be that we generally take a utilitarian approach and centre categories around the most common or striking values or types in our everyday experience, using whatever suitable delineating boundaries we can establish, often fuzzy. For example, we do this with 'age', and the electromagnetic spectrum, which are fairly simple one-dimensional categories; but we also do it with categories that are fuzzy in many dimensions, like 'life', 'alive', 'species', 'mind', etc. In my experience, intent is generally reserved for organisms that direct their behaviour in ways recognisably similar to our own, i.e. that have minds, some level of consciousness, and that 'think' (i.e. sophisticated information processing), using a specialised organ (brain) that can map the world and produce an internal model to predict and select future behaviour strategies based on the stored results of previous behaviour (learning, knowledge). Basing behaviour on a representation or abstraction of the world, on dynamic modelling, gives it a level of indirection, or '[URL='https://en.wikipedia.org/wiki/Aboutness']aboutness[/URL]' that is associated with intentionality. I'm not saying that this is how people decide whether to call an action intentional, but that it seems to me that when they use the concept literally, they're usually implicitly distinguishing that category. It seems to me that, to keep its semantic utility, the literal use of intent should be restricted to that category, and when the concept is used to describe behaviour outside it, it is anthropomorphic projection using the [URL='https://en.wikipedia.org/wiki/Intentional_stance']intentional stance[/URL], i.e. viewing behaviour in terms of mental properties that are not actually present, but that, as cognitively familiar terms, help us construct a meaningful or interesting narrative for it. But that is itself an example of anthropomorphic projection, the intentional stance writ large. An amoeba doesn't have the capacity for thought - a cognitive sense of self, or capacity for planning and forethought; if that level of cognitive capability was possible without a brain, we wouldn't need or have brains. This is why it's I think it's important to recognise when the intentional stance is used; e.g. a raindrop on the window isn't really [I]contemplating[/I], then [I]deciding,[/I] then [I]choosing[/I] the path of least resistance down the window, a thermostat doesn't really think "Is it warm enough for me to switch off yet?", then make a [I]decision [/I]to switch when it [I]thinks [/I]the temperature is just about right. Infants and children seem to be primed to interpret events in the world in terms of anthropomorphic narratives (i.e projected sentient agency), which is probably a by-product of innate predisposition to [URL='https://theness.com/neurologicablog/index.php/hyperactive-agency-detection/']Hyperactive Agency Detection[/URL], and their developing [URL='https://en.wikipedia.org/wiki/Theory_of_mind']theory of mind[/URL]. This tendency [URL='https://blogs.scientificamerican.com/thoughtful-animal/animating-anthropomorphism-giving-minds-to-geometric-shapes-video/']persists into adulthood[/URL], by which time the distinction between the reality and the anthropomorphic narrative is usually well understood; nevertheless, the anthropomorphic narrative is often the preferred way to interpret activity, and can be a source of entertainment or amusement (as in cartoons and movies of animals behaving like humans, and the predilection for dressing up pets). 'Chaos' occurs when the output of a system is sensitively dependent on its initial conditions; it's an exponential divergence in output or outcome for arbitrarily small differences in input or starting conditions, but it is deterministic. Given that amoebae use external chemical concentration gradients to detect food particles and organise their movement, and that their internals are fairly dynamic, it wouldn't surprise me if putting amoebae into a uniform environment, i.e. even lighting conditions, without chemical gradients, etc., would produce pseudo-random or chaotic movement, as any movement would result from small unpredictable variations in their internal conditions. Chaotic behaviour seems to me a rather [I]poor[/I] indicator of intent - all kinds of simple natural phenomena display chaotic behaviour, from [URL='https://en.wikipedia.org/wiki/Brownian_motion']Brownian motion[/URL] to the path of lightning strikes. It seems more likely that we should expect to see a relatively [I]low[/I] incidence of chaotic behaviour in living things, as goal-oriented behaviours are more likely to tend towards common paths or solutions to particular goals (although an injection of chaotic behaviour is sometimes used to make life difficult for predators). [/QUOTE]
Insert quotes…
Verification
Post reply
Forums
Discussion and Debate
Discussion and Debate
Physical & Life Sciences
Intent
Top
Bottom