"Why does HI lay such stress on gradients of well-being? From an ethical perspective, wouldn't a permanent maximum of bliss be better?"
A motivational system based entirely on heritable gradients of well-being is a less radical prospect than the abolition of motivation altogether. This is because hardwiring constant maximum bliss entails discarding the information-signalling role of the pleasure-pain axis completely - not just recalibrating its scale. Barring some extraordinarily advanced technology, uniformly happy beings will be out-reproduced. So for the foreseeable future, at any rate, encoding a physiological maximum of lifelong bliss is simply not an evolutionarily stable strategy. Then there's ideology to consider. If maximising gross cosmic happiness depends on (post-)humans embracing a classical utilitarian value system, it's presumably an unlikely scenario on that score too. Pluralist or perhaps quasi-utilitarian value systems are more sociologically plausible. Yet HI's (tentative) forecast that a motivational regime of gradients of bliss will be conserved indefinitely is itself no more than a conjecture. One counterargument is that choosing less fulfilling states of mind runs counter to the hedonic roots of our decision-making psychology itself. When mature technologies of emotional self-mastery become ubiquitous, it's uncertain who - if anyone - will really settle for what subjectively feels like an inferior option. What dial-settings will rational agents choose for their own mood-range when freed from the old Darwinian roulette? In practice, informed preference utilitarianism and classical utilitarianism tend to converge. Just possibly, the cumulative outcome of our choices may be the transcendence of traditional decision-making. As a slogan, "freedom to control one's emotions" invites readier assent than "freedom to enjoy limitless bliss". What's unclear is whether the ultimate cosmic outcome will be substantially different - or ethically, whether it ought to be so. Obviously care should be taken here to separate normative judgement from positive prediction. Certainly, billions of years of pan-galactic hedonism isn't quite what Jeremy Bentham had in mind when first enunciating the greatest happiness principle. A lawyer by training, Bentham had in mind institutional and legislative reform. Yet harnessing biotechnology to a classical utilitarian ethic dictates saturating the cosmos with blissful euphoria/positive value and then computationally sustaining this theoretical maximum indefinitely - whether in the form of discrete superminds or perhaps a Borg-like collective mind. The logic of "hedonistic" utilitarianism is inexorable, even if its premises can be challenged.
The issue of whether we should encode hedonic gradients or constant happiness should be distinguished from the related question of so-called "higher" versus "lower" pleasures i.e. the notional value of whatever we may be happy "about". Gradients of cerebral well-being (or ill-being) can certainly facilitate critical discernment, rational decision-making, and motivated behaviour. Yet as our rapidly evolving computer software attests, neither qualia nor an organic substrate are essential to this functional role. So as our integration with intelligent software increases, the "texture" of subjective dips of bliss may turn out to be functionally unnecessary for sentient organic life too. Tomorrow's technologies of fine-grained emotional control may enable early post-humans, for instance, to amplify their most treasured second-order desires for, say, cultural excellence, intellectual acumen and moral integrity while banishing the baser carnal passions. But after exploring the richest hedonic backdrop to whatever it is one most values - whether highbrow or lowbrow by today's lights - will anyone revert to hedonically impoverished states on discovering what they've been missing? Does our contemporary revulsion from crude wireheading, for instance, lie in the unvarying bliss that it yields - or merely its unedifying focus? Thus it's conceivable, as the Objection implies, that our distant descendants will enjoy some kind of ceaseless rapture - perhaps contemplating unimaginably sublime beauty or love or elegant mathematical equations. Or, less portentously, hilariously funny jokes. Naturally, these examples are purely illustrative, since post-humans may be imbued with kinds of blissful experience whose categories Homo sapiens can't name or conceive. Perhaps post-humans will be temperamentally meditative; perhaps dynamic. Perhaps they'll live in augmented organic virtual reality; or perhaps they'll live in designer VR paradises run on different bylaws from our presumptive basement. Perhaps they'll inherit a recognisable descendant of ordinary waking primate consciousness; or perhaps they'll live in unknown realms of utopian psychedelia. Unfortunately, our ignorance of the potential varieties of blissful experience contributes to the misconception that such well-being will necessarily be "thin" or unidimensional rather than diverse. But whatever the scenario, there's indeed no guarantee that a rational superintelligence will tolerate any decrements of well-being, information-signalling or otherwise.
The Objector's vision of unvarying bliss doesn't appeal to the dominant Western ethos. For the most part, modern capitalist societies prize innovation, creativity and change. So the prospect of a civilisation based (merely) on gradients of extreme well-being may be less unsettling than a future of constant bliss - though either condition is alien to Darwinian life. We associate permanence with stagnation; and passivity with low motivation and malaise. So any "static" vision fails to inspire. From a broader evolutionary perspective, self-propelled bodies exhibiting goal-directed behaviour arose early in the history of multicellular life on earth. This architecture has been strongly conserved over hundreds of millions of years. Looking ahead to an era when intelligent life has conquered raw suffering, and to an era when we can modulate our core emotions at will, enhanced hedonic gradients and/or their functional analogues may lead our post-human descendants, and/or our intelligent robots/cyborgs, to radiate and colonize every niche of the accessible multiverse within our light cone/galactic supercluster and intelligently re-engineer it. But what then? The (hypothetical) discipline of secular eschatology won't always be the idle fancy it seems at present. After we can effectively ring the changes within the finite state-space of matter and energy in our cosmic neighbourhood, which kinds of supersentience will be judged worth instantiating? To use a lame analogy, will we opt endlessly to replay mediocre games of chess or painting-by-numbers? Or confine ourselves to the state-space of perfection? Is status quo bias as irrational in post-Darwinian paradise as it is in Darwinian purgatory? On the Objector's "constant bliss" scenario, everything formerly unpleasant or mediocre - from avoidance of noxious stimuli to the mundane maintenance of the infrastructure of civilisation - will presumably have been computationally "offloaded" onto our intelligent machines/prostheses. Critically, selection pressure will no longer operate since post-humans will have occupied every possible niche and engineered themselves to have become effectively immortal. The old era of frenetic "action", the sound and fury of imperfect lives played out against a backdrop of restless discontent and scarcity economics, will belong to our animalistic ancestry. Even the transitional era defined by gradients of cerebral euphoria will have been left behind. Quite possibly the molecular signature of all valuable experience will have been identified; and its substrates amplified to the full. Indeed, given the pleasure principle plus advanced technology, an evolutionary trajectory to the presumed attractor of ideal states of sentience may be inescapable. Once the transition to grown-up consciousness is complete, the theoretical possibility of venturing outside this state-space may be even less likely than, say, our now deciding to revisit the lives of savages in caves. If and when intelligent life reaches cosmic superheaven, perhaps the baroque scaffolding that got us there will be kicked away. Eternal bliss needn't be orgasmic in the sense of lacking all intentional objects beyond itself; but presumably even this must be an open question. Either way, "timeless" bliss doesn't have to feel static. Mastery of the neurochemistry of time perception may allow each here-and-now to have a vast temporal depth, a rich internal dynamics, and subjectively to last an eternity. But perhaps speculations about the far future of cosmic consciousness are best avoided.
It should be stressed that all such wild post-Darwinian scenarios are remote - and vastly more speculative than the abolition of suffering or radical motivational enrichment. Hitherto in history, fitness-enhancing gradients of discontent have been the motor of progress - intellectually, socially, aesthetically, morally, personally. Most of the discontent endemic to the living world has indeed been unproductive; but not all of it. So harnessing the information-bearing role of its functional analogues - i.e. dips or anticipated dips of subjective well-being that still feel wonderful, but not sublime - is a more practical stopgap than encoding constant bliss. After all, we're barely on the eve of the reproductive revolution of designer babies, let alone an era of advanced paradise-engineering. In the near-to-medium term, recalibrating the genetic dial-settings that regulate hedonic tone is a less challenging bioengineering task than offloading everything to smart machines and replacing the old motivational and affective homeostatic control mechanisms of organic life completely. Gradient-surfing is also more ideologically realistic. Moreover even on the more conservative gradients-of-bliss scenario, any subjective "cost" of hedonically sub-optimal states i.e. information-signalling dips in well-being - is presumably acceptable to all but the most ardent utilitarian ideologues. Thus in future our hedonic baseline of mental health can still be richer than today's peak experiences. Assuming that the information-signalling role of gradients in well-being is indeed retained, any functional decrements of bliss can still be small. Even if the gradients are exceedingly subtle, there is no risk of a "Buridan's ass" scenario. [Buridan's ass was a mythical mediaeval equine which starved to death from indecision after being presented with the option of two equally appetising stacks of hay]. It's depressives who are prone to procrastinate; by contrast, happy people are typically decisive, extremely happy people more so. Indeed HI predicts that our immediate descendants at least will not be "passively", uniformly happy but hypermotivated, albeit on a much higher plateau of well-being than our current neural architecture can support. Enriching the reward centres of contemporary organic life will tend to heighten both its sense of purpose and purposeful behaviour - though to what end we don't know. Admittedly, this association of enhanced motivation with enhanced well-being may only be a contingent fact of our neural architecture - an accident of evolutionary history. The mesolimbic dopamine ("wanting") and mu opioid ("liking") neurotransmitter systems have co-evolved; their functional roles can in principle be disentangled. But again, a separation is scarcely imminent. (Post-)human agency still has a long future.
Depending on the strength of our bioconservative prejudice, gradients of adaptive well-being needn't be heritable. In principle, designer drugs, neurochip implants, nanobots, or autosomal gene therapy could achieve the same result - even within the constraints of a contemporary genome. But if our existing motivational system is defective, then it would seem cruel not to cure the pathology rather than transmit it to future generations. We wouldn't now consider it ethical deliberately to pass on genes for, say, a chronic pain syndrome on the grounds that our future pain-wracked offspring should be "free to choose" whether they wanted to be pain-free or not. Ethically, are our more pervasive syndromes of psychological malaise any different? Why shouldn't mental superhealth be heritable too?
How about the very long-term future? Normative judgements aside, will motivation in the traditional sense endure as long as sentient life itself? Could a future informational economy of mind based on gradients of bliss culminate in some sort of timeless cosmic paradise? Early in the 21st century, at any rate, this sort of question is probably too difficult to answer.
4.34 ...why the headlong rush? Let's wait until we have the wisdom to understand the implications of what we're doing... 4.35 ...the Simulation Argument suggests suffering can never be abolished...
E-mail Dave : firstname.lastname@example.org