In the next section, "Meaning-Making Within and Without Religion," I believe Lewis falls into the literalist trap on epistemology, while failing to consider the sociological implications altogether.
Chapter 10 is entitled "Where does this leave religion and spirituality?" as though he has explained a "this" which he may now treat as a jumping off point. While I would strongly disagree with this, since he has barely scratched the surface of the questions I consider crucial to his psychology-based investigation of the deep connections guiding meaning, I think I can fairly set out what he thinks he has arrived at.
Lewis seems ready to treat all belief in "alternative" or "supernatural" mechanisms of influence on the world as a definition of "religion", which of course accepts literalism at face value. He has made a strong case that such beliefs are based in psychological error and prejudice, and that morality is not derived from or dependent on any authoritative revelation by transcendent sources of understanding. In his introductory paragraphs to the chapter, he simplifies still further by stating his view as one that religion saw ultimate origins in terms of an Intelligent (and all-powerful) designer, that science has repeatedly shown other mechanisms of origin, so "where does that leave religion?" If it is no longer viable as bad science, in other words, is it valuable at all?
In this chapter he, as usual, takes on several related issues. He examines literalism, basing his discussion mostly on Karen Armstrong's understandings of the historical processes generating scripture and belief. He argues that efforts to salvage ever-more-remote roles for outside processes, mostly in the form of "teleology" in evolution, are misguided. He examines the question of compatibility between religion and science, IMO very badly. He then treats the question of "A continuing role for religion" in terms of modernistic conceptions of what is going on in religion, heavily emphasizing Armstrong's concept of "mythos" (but not, IMO, really understanding it) arguing, essentially that religion must ditch ("de-emphasize") supernaturalism and concentrate on moral teaching if it wants to remain socially relevant and influential on anyone except those dedicated to embracing literalist delusion. This may sound like the same line I tend to argue, but Lewis has inverted some key aspects of the perspective.
Working from Armstrong's cogent presentations, Lewis first reviews literalism and the origins of the scriptures in Abrahamic religions, arguing that the stories told about the origins of the religions do not hold up as accounts of their actual origins.
In one section on "still believing them to be the Word of God," he asserts that religions believe the best ideas were revealed in the past, while according to science, the best ideas will be discovered in the future. Leaving aside the dynamics of claiming that nuclear weapons, for example, are "the best ideas", this is a dichotomy I would argue is false. Both religion and science are engaged in improving our picture of the world, and the benefits of either and both depend on our choices, not on some beneficence or authority of either one.
In a section on "abstract" conceptualization of religious beliefs without God as author of revelation, he introduces the "apophatic" set of concepts about God, which tend to insist that we cannot know what God is like and that our efforts to define and understand God are self-defeating. He then puts Taoism in that category, raising questions of overly casual treatment. More to the point, he arrives at a term, "abstract theists" which refers to those who use non-anthropomorphic notions like "universal spirit" to think about God, arguing that these are all very well but the average religious person needs more specific and concrete conceptualizations.
He also asserts that abstract notions of God are "still characterized by the basic human habit of assuming that the universe possesses conscious agency or intention, that it exists a priori for some predefined reason, and that it is "seeking" to consummate some ultimate purpose." While his examples of such efforts, especially among people on the fringes of science, are good illustrations, I think he is trying too hard to impose his dichotomy. Just as those engaged in science don't usually have a very good conception of how religious thinking works, so it is not surprising that those engaged in religion don't usually have a very good conception of the way scientific understanding works. That doesn't mean one way of thinking has to be wrong and pernicious.
The topic of theodicy, addressing why an omnipotent and omni-benevolent God permits suffering and evil, seems to present a kind of test case for him, and he unsurprisingly reaches the conclusion that religion usually follows motivated reasoning to some logically untenable or morally blind "answer." After presenting the "natural teleology" views that somehow the universe is "trying to come to awareness of itself" or otherwise has a purposive guiding process underway, he reaches his own conclusion (after inviting the reader to draw his or her own) that inherent purpose in the universe simply represents misguided thinking.
Before pushing on, let me introduce the distinction that Armstrong uses, and that I think Lewis mishandles. She talks about the Greek distinction between "mythos", which includes literature and other story-telling, and "logos" or logical analysis. The point is not only that we have to regard each through the proper lens, looking at it for its correct purposes, but also that we need to understand the different ways language itself does its work in these different modes.
We are used to understanding the meaning of words and the operation of language through thinking about "logos" and with the tools of logos. Language is a way of passing on the contents of one mind, or set of understandings about the world, to another. In logos, the words have "correct" referents and any claims about the way the world works may be tested using logic, as well as evidence about the referents. This is vitally important for detecting deception, as Mr. Holmes and Dear Abby would be the first to assure us.
In mythos, by contrast, a word conveys much more (and less) than just its referent. It works in the right brain, integrating emotion with reality and creating associations between mood and image that cannot be quantified or formally tested. It can be both true and false at once, in its ability to connect reality with values. Consider the climactic final lines of "The Second Coming" (about terrorism roaring) by Yeats:
"And what rough beast, its hour come round at last,
Slouches towards Bethlehem to be born?"
The subtle mix of tone and mood, of tradition and jarring modernity, of Bethlehem the city of peace where a babe was found lying in a manger, swaddled in a blanket, with "slouches" and "rough beast" to evoke the resentful fury of the Irish Republicans and the cold predatory posture of the Sphinx, makes a potent potion no analysis can unmix. The use of language is more like music or dance than like law or science. To confuse the two is a natural, but dangerous, error.
The stories of Jesus calming the storm, of Peter getting out of the boat to walk on water, of healings and exorcisms, of the Spirit descending on Jesus like a dove, and, I would argue, of Resurrection, are mythos. They are meant to convey values and significance, not fact and causal sequence. A literalist, virtually always fixated on the issue of authority, cannot accept that distinction.
Lewis cannot seem to grasp this distinction, or its significance, despite having spent considerable time, it seems, reading Armstrong and considering her points. When he imposes a dichotomy between religion and science he is engaged in resolving issues of authority and thus will find the self-reinforcing errors of literalism almost inescapable. While his conclusions are viable enough, from a logos perspective, they are tone-deaf from a mythos perspective.
Before addressing where he should have gone, I would note that his effort to address Non-Overlapping Magisteria puts the inadequacy of his approach on full display. First, he fails to understand the idea, indicating (as usual in those who seek to dismiss it from a scientism perspective) he has not read the book. Second, he recklessly claims for science the mantle of appropriate way to investigate "moral values and meaning". Because, you know, nuclear weapons have shed a lot of light on those subjects. Just ask Kim Jong-Un.
The idea that a Sam Harris "scientific epistemology of values" is a tenable alternative to philosophy is howlingly arrogant. I think any modern person recognizes that science has a lot to tell us about the facts that are relevant to sorting issues of values. But the idea that science as a process and an enterprise should be given the reins for investigating issues of values requires a visceral rejection of normal human enterprise on the subject, which is to say, takes an emotional, value-driven position on the subject.
So let me lay out where I think this line of thought fails.
Let's start with the dichotomy, the idea that the issue is about competing "worldviews" and which shall be decisive. Would it strike you as sensible to pose the question of "which shall rule, the left brain or the right?" As Paul Simon put it, that requires that the right brain with its hopelessly emotional ways must "labor through the long and lonely night." ("Maybe I Think Too Much" is the name of the song.) Now, if I had to choose between a life without antibiotics and electricity or a life without music and art, I suspect I would choose the second. But what kind of fool thinks I have to choose? A literalist, is the answer.
Freed from the dilemmas of establishing religious authority, "abstract theists" have been discovering rich traditions of alternative influence. Instead of the imperial approach, where religious authorities tell you how to live and what you must believe, there are processes of mutual comfort, of "insurance" against the hard knocks of nature, of cooperation on raising children with a sense of security and affirmation, of listening, and sharing, and mutual tolerance for our shortcomings, and when it is needed, of forgiveness. These may all be wrapped up in the term "grace" that is dear to Christians. Other religions have similar conceptions (at its best, Buddhism is all about grace).
The task of integrating right-brain mythos traditions with left-brain logos processes is about bringing society to flourishing using grace and its soft, social power, not about pretending that material abundance justifies any imperial domination process.
So how do we integrate the two? In particular, what do we do about competing "worldviews"? To me it seems obvious that we should look at worldviews through a right-brain perspective. They are about integrating emotions and facts. So where we (holistically) see a misguided approach, an inappropriate set of connections, we can ask ourselves (left-brain style) what priorities are leading people to make the wrong connections, and how these priorities can be more sensibly satisfied. In some sense that is what Robert Tulip and the other geoengineers are trying to do, finding better ways to live with fossil fuel burning, but they are relying on technology to answer needs that are probably more spiritual. The answer to spiritual needs is almost never imposition of authority, and almost always grace.
Take a harder case within the sphere of social issues. Suppose that a broad class of society sees abortion rights and endorsement of alternative marriages as an insidious undermining of affirmation for child-raising and traditional motherhood. Exercising grace, we can influence entertainment and education to provide reassuring nurturance for such traditions. Education for parenting is a vastly underserved need, falling between the stools of "economic" justification for education (focused only on GDP) and traditional methods of passing on child-rearing skills. For every Murphy Brown having a baby as a single (affluent) mom, there should be three comedies or dramas exploring the complexities of family (including traditional family) in the confusing world most people inhabit. Fending off inappropriate authoritarianism, we can ask the difficult questions of why traditional families feel the need to have authority reinforcing them, and whether there might not be more useful ways society could help them achieve human goals.
Evangelicals in the U.S. are fighting a visceral, necessarily anti-rational rearguard action to defend their authority and power. Instead of vowing to defeat them, the broad middle of the country could simply meet them with grace, listening to what they want to see addressed, engaging in dialogue, and providing structures of support to reassure them that their fears do not need to lead them to disconnect from rationality and mutual trust.
Perhaps the most difficult part of that rear-guard action is the reaction to loss of male privilege. Seeing the world as a fight, a war, is the worldview that reinforces the shaming system against masculine vulnerability and spirituality. If we who believe in the mythos of sweet reason are to bring grace to that system and permit the peaceful future to be born, we need to think carefully about what emotional connections are being made and why, and to create social structures that can re-make them without the need for war.