The Process Machine: Using Prompts to Think With Yourself
Dialectical thinking refines ideas through opposition. Process thinking generates ideas through integration. Some problems need stress-testing. Others need something else entirely.
The Messaging Moat That Refused To “Click”
The whiteboard was a mess of crossed-out phrases.
I was working with a client on their messaging. They had a real moat, but it refused to behave like one.
They had genuine technical differentiation: architecture decisions that would take competitors years to replicate. A founder with twenty years of domain credibility whose name opened doors before he said a word. A contrarian thesis about where their industry was heading that looked crazy three years ago and was now proving correct. And unusual distribution: relationships and channels built over a decade that couldn’t be copied by hiring a sales team.
Each was defensible. The problem was that the client had all four.
Classic positioning doctrine says: pick one. The market can only hold one idea about you, so make it the sharpest possible idea. I’d internalized this. I had a toolkit for exactly this situation: dialectical stress-testing. Pit the options against each other. Steelman each angle. Find the one that survives.
So I ran the process. Four rounds.
Lead with technical differentiation? You sound like every startup claiming proprietary technology.
Lead with founder credibility? You undersell the product and when he’s not in the room, the messaging collapses.
Lead with the contrarian thesis? You attract intellectually curious people who want to discuss ideas, many of whom aren’t buyers.
Lead with distribution? That’s an operational advantage but not a positioning statement.
Four positions that couldn’t quite survive scrutiny. Every time I sharpened toward one, I misrepresented the others. Every time I tried to pick a winner, the messaging felt true-but-partial.
The Integration Problem
After the fourth round, I stepped back and looked at the whiteboard differently.
The moat wasn’t any single element. The moat was the combination.
A competitor could replicate any one piece. Given enough time and resources, the technical architecture is just engineering. Credible founders can be hired or acquired. Ideas spread. Relationships compound but aren’t impossible to build.
What a competitor couldn’t easily replicate was all four, integrated, reinforcing each other. The founder’s credibility gave the contrarian thesis weight. The thesis attracted buyers who could appreciate the technical differentiation. The differentiation justified premium pricing that funded the relationship-building that created the distribution advantage. The distribution generated case studies that reinforced the founder’s credibility.
The moat was a system. Isolating any element actively misrepresented the competitive position.
But here I was, with a toolkit designed to select winners, facing a problem that required integration.
The dialectical approach kept asking: “Which of these survives?”
The actual question was: “What emerges when these are held together?”
A Different Kind of Thinking
The previous essay in this series, The Dialectical Machine, offered a framework for using LLMs to think against yourself. Four modes of adversarial engagement designed to strengthen thinking through opposition. The assumption underneath that framework is that good thinking requires friction. Ideas that never meet resistance remain untested.
That assumption is true, but incomplete.
Some problems are selection problems. You have multiple options and need to determine which one survives scrutiny. Dialectical thinking excels here.
Other problems are integration problems. You have multiple valid elements and need to discover what becomes possible when they’re held together. Dialectical thinking fails here because it keeps looking for winners where the point is combination.
The messaging project was an integration problem. So is much of creative work and early-stage exploration when you don’t yet know what you’re looking for. It’s true of any situation where multiple perspectives are valid and the goal isn’t to pick one but to honor all of them in something new.
For these problems, opposition is the wrong mode.
The philosopher Alfred North Whitehead spent his career articulating what the alternative looks like. He called it process philosophy. Where dialectical thinking sees collision and resolution, Whitehead saw something else: entities taking each other in, incorporating rather than defeating, and creating novelty through integration rather than selection.
What follows is a framework for using LLMs in this process mode. Four ways to think with yourself rather than against yourself.
Whitehead’s Alternative: Process as Creative Advance
Alfred North Whitehead was a mathematician who became a philosopher. He collaborated with Bertrand Russell on Principia Mathematica, then spent the rest of his career developing an alternative to the Western philosophical tradition’s obsession with static substances and fixed categories.
His central insight was that reality isn’t made of things. It’s made of events, occasions, processes of becoming. Nothing simply is. Everything is always becoming.
This sounds abstract until you apply it to thinking itself. The dialectical model treats ideas as static positions that collide. Thesis meets antithesis. One survives, one dies, or they merge into synthesis. The movement is combative: collision, resolution, winner.
Whitehead offers a different model. Ideas don’t collide—they prehend each other. Prehension is his term for grasping, incorporating, taking-in. When an entity prehends another, it doesn’t defeat it. It takes it in, lets it become part of its own becoming. The other entity isn’t destroyed. It’s incorporated.
This changes things about how ideas can relate in interesting ways.
In dialectical thinking, differences are tensions to be resolved. In process thinking, differences are contrasts to be held.
Whitehead’s contrast is the simultaneous presence of differences that creates intensity and richness. The more contrasts held together harmoniously, the more value in the outcome. Harmony is the productive holding of difference, as opposed to absence of difference.
The process by which multiple inputs become one novel output, Whitehead called concrescence. The many become one, and are increased by one. Multiple elements don’t compromise or average out. They become a new actual entity that wasn’t contained in any of the inputs.
One more concept matters here: Whitehead saw propositions differently than the logical tradition. A proposition isn’t just a statement that’s true or false. It’s a lure for feeling—an invitation, an attraction toward possibility. Propositions invite consideration, open up spaces for exploration.
Translated into thinking practice:
Prehension means ideas can incorporate other ideas without defeating them. You don’t have to pick a winner. You can ask: what does this idea look like when it has fully taken in that one?
Contrast means differences can be held together productively rather than resolved prematurely. You can ask: what becomes visible when I refuse to collapse this tension?
Concrescence means genuine novelty can emerge from integration. You can ask: what new thing wants to exist that draws from all of these inputs?
Lures for feeling means propositions can invite rather than assert. You can ask: what questions attract the mind toward this territory rather than demanding conclusions about it?
These concepts translate into four modes of process prompting.
Mode 1: Mapping the Prehension
Prehension Mapping helps you see how ideas might incorporate each other without either one winning or losing. It’s the antidote to the forced-choice thinking that dialectic can produce.
When to Use This
Use this when you have two or more positions, frameworks, or perspectives that seem in tension, and you suspect that the tension might be productive rather than something to resolve. When you’ve tried to pick a winner and something valuable keeps getting lost.
What You Bring to This Dialogue
Two or more positions that seem to compete. These could be strategic options, theoretical frameworks, creative directions, or perspectives from different stakeholders. The key is that each has genuine merit and you’re not looking to eliminate any of them.
Clarity about what each position values. Before you start, articulate what each position is protecting or prioritizing. Often positions that seem opposed are actually protecting different values that both matter.
Genuine openness to transformation. Prehension doesn’t leave the original positions intact. When Position A fully takes in Position B, it becomes something new. You need to be willing to let your original framings transform.
The Dialogue Structure
Begin with this framing:
I have multiple positions that seem to be in tension, but I suspect forcing a choice between them loses something important. I want to explore how each might incorporate the others rather than defeat them.
Here are the positions:
Position A: [State clearly, including what it values or protects]
Position B: [State clearly, including what it values or protects]
[Position C, etc. if relevant]
Help me explore prehension—how each position might "take in" the others:
1. What does Position A look like when it has fully incorporated the concerns and insights of Position B? Not compromised—transformed through integration.
2. What does Position B look like when it has fully incorporated Position A?
3. What new position might emerge that isn't reducible to either original, but honors what both were protecting?
Don't look for a winner. Look for what becomes possible when these are held together.After the LLM responds, push deeper on the integrations that feel generative:
The integration you described for [specific point] is interesting. Help me develop it further. What does this integrated position look like in practice? What would it say that neither original position would have said alone?What Good Prehension Looks Like
You’ll know the process is working when you start seeing possibilities that weren’t visible from any single original position. The integrated view should feel like genuine emergence (something new) rather than compromise or averaging.
Sometimes you’ll discover that positions you thought were opposed were actually describing different facets of the same underlying reality. Sometimes you’ll find that Position A’s weaknesses are exactly where Position B is strong, and vice versa.
The goal is to discover what becomes possible when you stop forcing a choice.
If the LLM offers resolution too quickly, push back:
That feels like collapsing the tension rather than holding it. I’m not looking for a middle ground or a synthesis yet. Help me see what’s generative about keeping both poles alive. What becomes possible because the tension exists?What Good Contrast Holding Looks Like
Productive contrast holding often produces reframes. You start seeing the tension differently.
You’ll know it’s working when you feel less urgency to resolve. The discomfort of holding both becomes interesting rather than intolerable.
Sometimes you’ll eventually resolve the tension but from a place of understanding rather than escape. Other times you’ll realize that holding the tension permanently is the right approach, that resolution would actually be a mistake.
Mode 3: Facilitating the Concrescence
Concrescence Facilitation helps you discover what new thing might emerge from multiple inputs. It’s not about combining or averaging, which is pretty straightforward. You’re looking for a genuine emergence of something new.
When to Use This
When you have multiple threads, influences, constraints, or desires and you want to see what novel integration might be possible. Or when you’re doing creative work and want to discover what wants to exist rather than forcing a predetermined outcome.
What You Bring to This Dialogue
Multiple diverse inputs. These could be ideas, influences, constraints, goals, aesthetic preferences; anything that will shape the outcome. The more diverse the inputs, the more potential for genuine novelty.
A genuine question rather than a predetermined destination. Concrescence facilitation works best when you’re truly uncertain what should emerge. If you already know what you want, you’re not doing integration.
Patience for the process. Genuine emergence takes time. The first thing that appears isn’t usually the final form. You need to be willing to let the integration develop through multiple iterations.
The Dialogue Structure
Begin with this framing:
I have multiple inputs that need to become something new. I’m not looking for compromise or averaging—I want to discover what novel thing might emerge from their integration.
Here are my inputs:
[List each input—ideas, constraints, influences, requirements, etc. Be specific about what each contributes or demands]
Help me facilitate concrescence—the process of these many becoming one:
1. What patterns or resonances do you notice across these inputs? Where do they naturally amplify each other?
2. What tensions exist between them? (Don’t resolve these yet—note them.)
3. Given everything here, what new thing wants to exist that isn’t simply a combination of the inputs but a genuine emergence from them?
4. Describe this emergent possibility in enough detail that I can evaluate whether it honors what each input was contributing.
The goal isn’t to satisfy all inputs equally. It’s to discover what becomes possible when they’re all present in the process of becoming.After the LLM offers an initial emergence, develop it:
That’s interesting. Help me develop [specific emergent possibility] further. What does it look like fully realized? How does it honor [specific input] and [specific input] simultaneously—not by compromising but by transcending?What Good Concrescence Looks Like
Genuine emergence surprises you. It’s not what you would have designed if you’d started from scratch, and it’s not what any single input would have produced alone. It feels like a thing that wants to exist rather than a Frankenstein assembly of parts.
You’ll know it’s working when the result seems obvious in retrospect even though you couldn’t have predicted it. When you look at it and think: yes, this is what all those inputs were pointing toward, even though none of them said it directly.
Sometimes concrescence produces something that doesn’t satisfy any single input perfectly but is more valuable than perfect satisfaction of any one would have been. The new thing has qualities that weren’t in any of the inputs.
Mode 4: Crafting the Lure
Lure Crafting helps you create propositions that invite exploration. It’s the mode for opening up possibility space.
When to Use This
When you want to attract attention and interest toward a territory rather than make claims about it. Or you’re at the beginning of exploration and need new and better, more generative questions rather than answers.
What You Bring to This Dialogue
A territory you want to explore. This could be a problem space, a topic, a creative direction, or a question you’re circling. You only need a sense of where you want to look.
Willingness to stay in question mode. Lure crafting is about generating invitations, not conclusions. If you find yourself wanting to answer the questions you’re crafting, slow down. The lures are the point.
Attention to what attracts. Notice which framings make you want to explore further. That attraction is a signal. Lures that don’t attract aren’t working.
The Dialogue Structure
Begin with this framing:
I want to explore a territory, but I’m not ready for conclusions. I want to craft lures—invitations that attract the mind toward this space rather than assertions that close it down.
Here’s the territory I want to explore:
[Describe the problem space, topic, or direction. Share what draws you to it, what you’re curious about, what you sense might be there.]
Help me craft lures for feeling—propositions that invite rather than assert:
1. What questions make this territory feel alive and worth exploring? Not questions with easy answers—questions that pull you in.
2. What framings make the unexplored possibilities here feel attractive? How might someone fall in love with this problem space?
3. What invitations could I extend—to myself or others—that would open up exploration rather than demand conclusions?
4. What would make someone want to spend time here, not because they have to but because they’re genuinely drawn?
Help me make this territory magnetic.Evaluate the lures by their pull:
The question about [specific framing] actually makes me want to explore. Help me develop more lures in that vein. What adjacent questions or invitations have the same quality of pull?What Good Lures Look Like
Effective lures create attraction. You read them and want to think about them more.
Good lures often have a quality of incompleteness that’s generative. They suggest that there’s something here worth finding, without telling you what it is.
You’ll know your lures are working when you find yourself thinking about them when you’re not trying to.
When to Use Which: Dialectic vs. Process
The two machines are complements, not competitors or even mutually exclusive.
Use the Dialectical Machine when:
You need to stress-test a specific claim before going public with it
You’re deciding between options and need to find the one that survives scrutiny
You want to eliminate weak thinking before investing more in it
You need precision and commitment
The problem is fundamentally a selection problem
Use the Process Machine when:
You’re exploring and don’t yet know what you’re looking for
Multiple perspectives are valid and the goal is integration rather than selection
You’re doing creative work where novelty matters
The tension between options seems productive rather than problematic
The problem is fundamentally an integration problem
Use them in sequence:
Process prompting to explore and generate. Dialectical prompting to test and refine what emerges.
Or: Dialectical prompting to stress-test an initial position. Process prompting to integrate what you learned into something new. Dialectical prompting again to test the integration.
The point is to recognize which problem you’re facing and reach for the right mode.
The Moat Resolved
After developing these process modes, I returned to the client’s messaging problem.
Instead of asking which positioning angle survived, I asked what emerged when all four were held together. The technical differentiation and founder credibility and contrarian thesis and distribution advantage—what did they prehend from each other? What contrasts between them were worth holding rather than resolving? What new positioning might concresce from all four that wasn’t reducible to any single one?
What emerged was a message about compounding advantages. The positioning wasn’t a was the explicit claim that this company had built something that reinforced itself. Technical excellence attracted talent that expanded distribution that generated proof points that validated the thesis that elevated the founder’s credibility that attracted more technical excellence.
The dialectical machine refines what exists. The process machine discovers what might become. Both are necessary. Neither is sufficient alone.
Talk again soon,
Samuel Woods
The Bionic Writer