Alison Sinclair

In an exchange on a listserver I am on, the question of writing lessons learned along the way came up. This was my list . . .

• Published novels are the finished product: one never sees the messes, failures and train-wrecks on the way, so one is completely misled as to how easy certain things are to execute. The downside of a diet of the best is that the emerging writer can become inadvertently overambitious and try things that are too difficult for them.
• I did two dumb things and two smart thing in my first novel. Dumb things (ie, things I wasn’t developed enough to do): writing a quest novel, and using that past-present structure that Ursula Le Guin made work so beautifully in Dispossessed. I didn’t realize until a year or so after Legacies came out where I’d got it from, and why I was so wedded to it. The sort-of-quest structure is difficult to pull off because it doesn’t innately have a strong narrative drive behind it. Smart things I did: having a single viewpoint, and having a character I had deliberately written as attentive and extremely perceptive. Sometimes, wrestling with the need to convey something essential via a viewpoint character for whom it’s not in character to notice that, I miss Lian.
• Certain plots are more bomb-proof than others – they carry their own structure and drive with them. Blueheart‘s initial plot is a mystery, and once I’d got that – the dead body in the ocean – it found its shape quite quickly, carried along by the central question of who and how. By midway through the book the reader actually knew everything, and it turned into a political novel, but by then the central conflict was established and on its way to the climax. I did myself an inadvertent favour, there.
• Quest plots – frequently the first plot an SF&F writer tries – are not as easy as they look: certain choices have to be made to ensure the quest plot gets and keeps its narrative drive and doesn’t become picaresque (a right-on editorial comment about an early draft of Legacies). If I were writing a quest, even now, I’d make sure that what was being sought and who was seeking it were established in the first chapter, and not lose sight of that for a moment. I’m still not sure enough in my plotting to do the young man/woman goes off all unknowing and find his/her destiny on the way. I was unwittingly smart enough to have the quest front and center in the beginning of Legacies’ frontstory, interspersing it with the interleaved backstory in which Lian had to find his mission.
• Passive, reflective characters fall under the heading of Advanced Work. Again, writers have pulled off the reluctant hero wonderfully, but life is much easier if a character wants something and goes after it. Lian climbing over the wall, throwing himself into the path of Lara and Rathla and the story itself, was a wonderfully liberating moment for me.
• Sometimes the writer just has to give up and do what’s obvious – usually because they’ve set themselves up that way. In one of my unpublished novels I was resisting a particular idea because it seemed too obvious. When I finally accepted that it had to be that way, a whole lot of other problems were suddenly solved, because my characters’ repugnance (they didn’t like the idea any more than I did) prompted them to actions that led directly to the showdown. Moral: It’s a bad idea for the writer to argue with their own story.
• Even after (almost) 9.5 novels, I still don’t get control of the plot until my second draft (or later). I’ve just had to do a massive overhaul to keep two of my main characters on the scene for a major action setpiece (this was Shadowborn). I also had difficulties setting up a crucial event in that conflict, because I needed not to surprise the reader, but I knew that if one of the characters knew about it, it would be out of character for him to leave. So overhaul. And it works. So. Much. Better. Moral of the story: keep the viewpoints where the action is. As long as the action is essential to the plot.
• If I reach the end of my first draft, and it isn’t right (usual metaphor: large plate of spaghetti, stands slithering over the sides), I start cutting. I usually have a fixed idea of the endpoint from fairly early on in the novel, and I reshape the novel to line up with the end. I cut out everything that that isn’t related to the end. Then I put in everything that’s missing.
• On the other hand, all the scenes that end up on the cutting-room floor mean that by the time I get the scene I need, it practically writes itself because all the decisions are made and I have the characters rounded out. Ibsen described his growing familiarity with his characters through successive drafts. In the first draft he knew them as if he had met them on a train (‘One has chattered about this and that’). By the second draft he might have spent a month at a spa with them (‘I have discovered the fundamentals’). By the third draft, he knew them thoroughly (‘as I see them now, I shall always see them’).
• I try to obey Chekov’s Law (‘One must not put a loaded rifle on the stage if no one is thinking of firing it.’), which usually means I have to round up a certain amount of unused artillery during revisions. One of the downsides of writing a trilogy is that once Darkborn was committed to press, I was committed to firing off the guns lying around. Twelve of them, when I did the inventory in my notebook. I was delighted when I found a way to get four to pop off at once in the archduke’s breakfast.

Cross-posted from Kayak Yak . . . Drug discovery from marine sources is an active area of research, and several drugs of marine origin have already reached regular clinical use. (There’s a whole journal dedicated to Marine Drugs. Open access, too). For quite some time, I’ve had the notion of doing a series of posts about these drugs, where they come from and what they do, and I’ve been tinkering with this first entry for about as long, trying to balance length with bio-geekery. So here goes.

In 2004, the pain-killing drug ziconotide (Prialt) was approved for marketing by the US FDA, some twenty years after Balmedro Olivera and Lourdes Cruz set out to find out how the sting from a 10 cm poisonous marine snail, Conus geographus, could kill a human many times its weight. C geographus (picture from Wikimedia commons) is a member of the cone snail family, so named for their distinctively shaped, colourful shells. C. geographus takes its name from its map-like shell pattern (photo from Wikimedia commons). It lives below the low tide mark in pockets of sand near the edges of coral reefs and atolls; on a map its distribution tracks a wide ribbon from Madagascar around the edges of the Indian Ocean, down through Indonesia, around the North coast of Australia and up the islands and atolls of the Pacific. To the biologist, it and its fellows are known as Class Gastropoda, Order Neogastropoda, Superfamily Toxoglossa, Family Conidae, while naturalists over the centuries have named the Apothecary Cone, Astrologers Cone, Hebrew Cone and Emaciated Cone. The dinosaurs were already extinct when the first cone shell was pressed into the fossil record, some 55 million years ago, but they now form a family around 700 species strong, which inhabits warm and tepid shallows around the globe. The more common species can be had for a pleasure-dive in reef waters, or a stroll along the ocean, while particularly rare and handsome specimens have been auctioned for more than the price of a painting (in 1798) or a family saloon car (1960). Occasionally, their price has been a life: 30 or so people are known to have died of cone snail stings, mainly from Conus geographus.

Cone snails’ dietary aspirations might seem, on first blush, overambitious: they are carnivorous, with a taste for worms, other molluscs … and fish. However, per Ecclestiates, the race is not to the swift . . . Cone snails generally use one of two hunting strategies, harpoon or net, and their secret weapon is poison, a cocktail of venoms tailored to hunting style and prey. The cocktails contains 50-200 individual toxins, and vary between species, so that there are an estimated 25 000 plus toxins across all the known cone snail species. The venom of the harpoon-wielding snail Conus purpurascens contains a mixture of fast-acting toxins that produce a nerve paralysis, and slower acting toxins that produce muscular paralysis. When a fish comes within range, the snail jabs at it with a venom-filled tooth held on the end of a proboscis. A stung fish can be paralyzed within two seconds, its body rigid and its fins standing out as though shocked. One toxin jams open the sodium channel involved in the propagation of the nerve impulse [footnote i], a channel that ordinarily would close immediately to allow the membrane to reset itself. Another jams closed those potassium channels which normally would open to quench the depolarization. The membrane depolarizes, rapidly and completely, and nerve conduction stops. Even as this happens, a second, slower-acting set of toxins, acting on calcium channels, starts to paralyze the fish’s muscles. They block sites to which signalling neurotransmitters bind, and sodium channels which would open in muscle contraction. Net-wielding snails, like Conus geographicus favour muscle paralytic toxins; since they first engulf their prey and then sting it, they can afford the slightly slower onset poison.

Contoxins are very small proteins, 10-35 amino acids long, and at their length would normally be a tumbling mixture of floppy conformations in solution rather than a fixed protein fold—Proteins depend on their ability to hold conformation to function. Conotoxins, however, take advantage of a property of the amino acid cysteine. Under the proper conditions, two cysteines in a peptide chain will link to each other, bringing their respective pieces of peptide chain into alignment. Conopeptides, small as they are, each have two or three pairs of cysteines, which cross-link to create a tight little package (3d structure of Ziconotide as the August 2006 molecule of the month at 3dchem). Since they are tightly folded, conotoxins waste neither time nor energy shifting into the right shape to bind to their target. There are two measures of quality of any interaction between molecules: how well a molecule discriminates between its own and all other binding sites, and how strongly it binds. By those measures, conotoxins are finely tuned, with certain conotoxins able to select between nerve cell sodium channels and muscle cell sodium channels, and others able to pick and chose between subtypes of calcium channel. Contoxins are several times more selective than peptides from snake and scorpion venoms. A measure of the strength of binding is the dissociation constant, a ratio of the amount of unbound toxins to the amount of bound toxin at a giving concentration; for the conotoxins those are of the order of 10(-9) or 0.000000001, or for every unbound toxin molecule, there are one trillion bound.

Drugs are often limited in their usefulness by side effects, some of which result from binding to molecular sites other than the target sites. For that reason, the conotoxin peptides, all 25 000-odd, with their strong, specific binding, are of great interest to scientists and several companies have investigated cone shell toxins as a source of drug candidates. Ziconotide (aka Prialt, from Elan Pharmaceuticals) is the first conotoxin-derived drug to pass successfully through all the stages of clinical drug development. It is a synthetic omega (calcium-channel binding) conotoxin from Conus magus, the Magician cone, which binds specifically to (calcium) channels in nerves in the spine which carry pain, and has proved effective in relieving pain for patients with intractable severe pain who either do not respond to or cannot tolerate other drugs. It’s not an opioid, therefore doesn’t produce tolerance, and can be combined with other drugs. But it’s still far from ideal: it has to be administered by intrathecal injection (directly into the fluid around the spine), which means it has to be given by an anaesthetist or by an implanted pump. The dose has to be increased slowly to decrease the risk of side effects, the onset of effect slow, response to dose changes is laggardly, and it can produce severe neurological and psychiatric side effects (Williams, 2008 PubMed abstract; Schmidtko, 2010 PubMed abstract).

To date, no other conotoxin-derived drugs have made it through clinical testing and into clinical use. Olivera in 2006 listed five in Phase I (first human) testing, mainly for pain, and their status in 2011 gives a snapshot of the vicissitudes of early phase drug development and the precarious life of small biotechnology companies: three of the companies involved appear to have either gone under or moved away from conotoxin development, and I am hard put to find evidence of progress on four of the compounds. The fifth (Xen-2174), which inhibits the uptake of the neurotransmitter norepinephrine, is in Phase II trials. Nevertheless, investigation continues; there are at least 24996 more conotoxins to go . . . A conotoxin derivative that can be taken orally has been developed. Conotoxins are being studied for their potential to protect brain tissue in stroke and heart muscle in heart attack, where part of the damage is known to be caused by uncontrolled leaks of ions across membranes [Twede et al, 2009 Full text]. And work with conopeptides has also identified other pathways involved in severe pain, leading to the development of non-conopeptide drugs directed at these pathways.

Footnotes and references.

[i] An aside on nerve conduction. It is orchestrated by the opening and closing of protein pores in the cell membrane of the neuron, which pass, according to their filter characteristics, sodium, potassium, or calcium ions. The sequence proceeds as follows: a trigger signal arrives, whether an electrical signal from another neuron, or a chemical signal. The membrane is held at a resting potential, a static voltage, with an excess of sodium outside the cell and potassium inside. When the impulse is triggered, sodium channels open, and sodium flows into the cell. The membrane depolarizes, losing its charge. Once the depolarization proceeds to a certain point, the sodium channels close, and potassium channels open, quenching the depolarization and allowing the membrane to reset to receive the next impulse. As each patch of nerve membrane is depolarized, it triggers depolarization of the next downstream; thus the impulse travels.

• Olivera BM. E.E. Just Lecture, 1996. Conus venom peptides, receptor and ion channel targets, and drug design: 50 million years of neuropharmacology. Mol. Biol. Cell. 1997 Nov;8(11):2101-2109.
• Olivera BM. Conus peptides: biodiversity-based discovery and exogenomics. J. Biol. Chem. 2006 Oct 20;281(42):31173-31177.
• Terlau H, Olivera BM. Conus venoms: a rich source of novel ion channel-targeted peptides. Physiol. Rev. 2004 Jan;84(1):41-68.
• Wikipedia on Conus species, Ziconotide, ion channel.
• ConoServer (peptide sequence and reference database).

As a R learner programmer, it took me unconscionably long to work out how to use polygon to shade under and between curves, despite searches of the R manual and R-help – they just didn’t start far enough back. So, for anyone else scratching his or her head over polygon (and so I can find it again when I forget how it’s done), here are the series of steps I went through to figure it out.

The function takes in an x vector and a y vector, defining a set of coordinates that, in order, taken in order trace around the area to be shaded. Thus for a set of points 1-10, defined individually as x.1, y.1 to x.10, y.10,

x <- c(x.1,x.2,x.3,x.4,x.5,x.6,x.7,x.8,x.9,x.10)
y <- c(y.1,y.2,y.3,y.4,y.5,y.6,y.7,y.8,x.9,y.10)

The area inside these points is shaded by

polygon(x,y,col=gray(0.8))

To apply this to two curves, both normal distributions between -3 and 3, one half the height of the other

x <- seq(-3,3,0.01)
y1 <- dnorm(x,0,1)
y2 <- 0.5*dnorm(x,0,1)
plot(x,y1,type="l",bty="L",xlab="X",ylab="dnorm(X)")
points(x,y2,type="l",col="red")
polygon(c(x,rev(x)),c(y2,rev(y1)),col="skyblue")

The first half of the x-vector in the polygon is just the values of x itself, corresponding to the part of the polygon that is tracing out the upper curve along increasing values of x. The second part for of the x-vector in the polygon is the reverse of x, corresponding to the part of the polygon that is tracing out the lower curve along decreasing values of x. The first part of the y-vector is the y values of the upper curve, and the second part of the y-vector is the y values of the lower part of the curve.

That shaded the area between the curves along the full plotted range.

To shade only a defined portion, say the area from x=-2 to x=1.

x <- seq(-3,3,0.01)
y1 <- dnorm(x,0,1)
y2 <- 0.5*dnorm(x,0,1)
x.shade <- seq(-2,1,0.01)
polygon(c(x.shade,rev(x.shade)),c(dnorm(x.shade,0,1),0.5*dnorm(rev(x.shade),0,1)),col="yellow")

To shade the same defined portion as a gradient, from red to yellow (from the built-in heat.colors palette):

x <- seq(-3,3,0.01)
y1 <- dnorm(x,0,1)
y2 <- 0.5*dnorm(x,0,1)
x.shade <- seq(-2,1,0.01)
par(oma=c(1,1,1,1),cex=0.7)
plot(x,y1,type="l",bty="L",xlab="X",ylab="dnorm(X)")
points(x,y2,type="l",col="gray")
l <- length(x.shade)
color <- heat.colors(l)
for (i in 1:l)
{
polygon(c(x.shade[i],rev(x.shade[i])),c(dnorm(x.shade[i],0,1),
0.5*dnorm(rev(x.shade[i]),0,1)),border=color[i],col=NA)
}

This draws a succession of individual polygons between the curves, adjusting the color along the gradient as it goes. (Note that the loop above is for i in one to letter-L).

I probably don’t have to explain Jo Walton’s singular new novel, Among Others, by now. If I do, in brief, it is the journal and reading diary of a fifteen-year-old girl who, after a series of traumatic events that left her twin sister dead and herself lame, has found a physical refuge in a girl’s boarding school, and a spiritual refuge in books, particularly SF. Which makes it sound mundane, except that Mor speaks to fairies—and not the Disneyfied, sentimentalized version, but the original wild spirits of wood, earth, and stone—works magic, and she and her sister sacrificed themselves to save the world. This is about the aftermath. It’s set in England in 1979-1980, and as someone who fell into SF while a teenager at a girl’s school in Scotland in the 70s, I really looked forward to the evocation of the time and the place.

I got that, and more. One of the unique aspects about the book was how well it portrays a young mind bursting out in all directions. I’d largely forgotten about that experience of intellectual flowering, of being in possession of an adult vocabulary and intellectual capacity, not to mention the toolkit that comes with a decent education, and being let loose with a fistful of library tickets to go romping through the best works of kindred and strange minds. I don’t think I ever found that experience—which has to be one of the best parts of adolescence, up there with the creative experimentation that goes with all those discoveries—portrayed in mainstream fiction—and how I snickered when when I read Mor’s caustic comments on Teen Problem Novels, because that is exactly what I thought, even then [i]. SF was a wonderful liberation from the mandated dreariness of adolescence. (I wonder if the experience gets encoded in SF and fantasy in the form of the emergence of psionic powers or magic . . . a topic for another time.)

The portrait of a character and a mind being formed by reading also made something go click in a way that hadn’t before: Reading is experience, as opposed to being a way of avoiding experience, or an inadequate replacement for it, a cultural assumption that I’d accepted (though not without resistance) for years [ii]. And because Mor absorbs her reading into her experience, on a number of occasions she simply says, “Oh, that”, and carries on, proceeding by the map her reading has laid out. Which in a couple of instances made me wince, and in others, laugh—pity the teenage lout who encounters a girl versed in Heinlein. Well, that was a laugh and wince, in sympathy.

And what about the magic? For myself as an SF/F reader, reading by a SF/F protocol, there’s no doubt: there’s magic. I like the magic, the subtlety of it, the way it merely leans on the possible. Having bounced off the endings of fantasy novels aimed at young readers (Silver on the Tree, The Last Battle), I like Mor’s explanation as to why she thinks that she will do less magic as she matures. Yet the very subtlety of the magic, makes it, as Mor says herself more than once, “deniable”. Mor has the hallmarks of reliability: she’s not a “fanciful” person; she might see fairies and ghosts, sense and work magic, yet she likes chemistry and physics, and would like maths if it would but like her back. She self-consciously challenges the putative reader’s skepticism only once, at the end of the first section of the book, unlike most first person unreliable narrators, who do so repeatedly. She’s prickly—I suspect Mor and (prickly!) fifteen-year-old me might not have got on—and sometimes brusque in her judgements, but deeply grounded and thoughtful, and has moved far beyond a self-centered view of right and wrong. Nevertheless, switch off the SF reader’s protocol, take a quarter turn, and consider the book from that angle, disbelieving in the magic, and the book still works as a study of a young woman’s imaginative response to loss. Which is another unique, and very neat, thing about it.

Footnotes

[i] I decided Teen Problem Novels were a product of cultural reaction against the teenage years of the the baby boom. Twentysomething boomers were embarrassed by adolescence, the rest of the culture was burned out on it, and nobody had anything good to say about it.

[ii] A few years after I discovered SF, I discovered feminism, and Joanna Russ provided me with a workable explanation as to the whole experience issue, which had preoccupied at least two generations of women writers before me (Mansfield, Plath, Russ herself): setting up ‘experience’ as a prerequisite, and certain kinds of experience at that, was another strategy that condemned women writers to insignificance. Worked for me for years, but I like this one better.

Cross-posted from Kayak-Yak.

Once upon a time in Brentwood Bay, while drifting over rocks studded with orange and purple starfish, and past huddles of starfish in crevasses at the waterline, it occurred to me to wonder why they were these colours, that purple, in particular. The starfish in question were the ochre star, Pisaster ochraceus, and the answer, after intermittent and desultory trawling through the web and the scientific literature, turned out to be (a) carotenoids and (b) maybe what they eat.

The Royal British Columbia Museum Handbook Sea Stars of British Columbia, Southeast Alaska, and Puget Sound, told me a lot about the anatomy, hunting and mating behaviour, but does not account for the colours: P ochraceus is the most common intertidal sea star, with territory from Prince William Sound, Alaska, to Cedros Island, Baja California (lucky it!), and from the intertidal zone to nearly 100 m undersea. It likes rocky shores, waves and currents. I’ve seen plenty in the Broken Islands, the Gulf Islands, and around Saanich Penninsula. P ochraceus eats mussels, barnacles, limpets, and snails. It is the paradigm of a “keystone species” in that its presence and predation significantly affect the numbers and distribution of other species, especially the California mussel, Mytilus californianus; in the absence of P ochraceus, M californianus takes over the beach. Pisaster spawn in May to July, releasing millions of eggs, which turn into larvae, first floating free in the plankton and then (those that survive) attaching themselves and turning into juvenile sea stars. Juveniles grow to adult size and maturity over about 5 years. Larval P ochraceus have a chemical defense that induces filter-feeders to spit them out (got to look that up). The only known predators of adult sea stars are seagulls and sea otters.

Harley et al, 2006 (full text available) looking at the colour variation, note in their introduction that “at least two caroteinoid pigments mytiloxantin and astaxanthin, sequestered in the aboral surface, produce these colors in Pisaster and other asteroids.” Aboral is the upper side side of the sea star, and starfish belong to the Class Asteroidea, under the Phylum Echinodermata. Caroteinoids as a chemical class are named after their best known member, the yellow pigment in carrots, and have in common a long carbon backbone with many concatenated double bonds which generally absorb light at the blue end of the spectrum, hence the orange colour. Mytiloxanthin was named after M californianus, part of P ochraceous’ preferred diet, from which it was first isolated, so it was assumed to be dietary in origin. Astaxanthin arises through “several distinct metabolic pathways”, and is orange. I’m still not sure from my reading what the pigment behind the purple is, though reading descriptions of 1940s-style chromatography makes me oddly nostalgic for undergraduate chemistry.

However, knowing the pigments doesn’t explain why individual starfish should be orange, ochre, brown, or purple, or why starfish on an exposed, wave-beaten rocky coast like the west coast of Vancouver Island should be predominately orange (6-28%) and brown (68-90%), while those in the sheltered waters of the South St Georgia strait should be almost entirely that brilliant purple so familiar on our paddles (95% in the samples collected by Harley). The answer is apparently not genetic: DNA studies don’t suggest that the populations sampled (from Alaska to California, with lots of attention to Puget Sound) are isolated from each other, and conversely do suggest that there is flow of genetic material between them. It’s not apparently to do with wave action, inasmuch as scientists have been able to reproduce in the lab the difference between turbulent water and calm. It may be dietary, in that the distribution of colours correlated with the pattern of prey: in the more exposed waters (where purple starfish are in the minority), P ochraceus preferentially eat M californianus, the big California mussel, whereas M calfornianus is uncommon to absent in interior waters (where purple starfish are in the majority), and the Pisaster there tend to prey on barnacles and bay mussels. So, eats purple mussels -> orange; doesn’t eat purple mussels -> purple. Hmm. And that still doesn’t explain why purple and orange starfish could be found within yards of each other. Another paper by Raymondi et al, 2007 (only abstract) found that the frequency of orange in a population was constant with latitude, but tends to increase with the size of the individuals in that population. So all is not quite explained.

References

• Harley CDG, Pankey MS, Wares JP, Grosberg RK, Wonham MJ. Color Polymorphism and Genetic Structure in the Sea Star Pisaster ochraceus. Biol Bull. 2006 Dec 1;211(3):248-262. And here’s marine biologist Christopher Mah (full name from his Twitter feed), on the Echinoblog, with a crisp and colourful synopsis, complete with photos and diagrams; if I hadn’t written a chunk of this entry while back before I found his entry, I’d just have said, go there!
• Lambert P. Sea Stars of British Columbia, Southeast Alaska, and Puget Sound. 2nd ed. UBC Press; 2000.
• Raimondi PT, Sagarin RD, Ambrose RF, Bell C, George M, Lee SF, et al. Consistent Frequency of Color Morphs in the Sea Star Pisaster ochraceus (Echinodermata:Asteriidae) across Open-Coast Habitats in the Northeastern Pacific. Pacific Science. 2007 4;61(2):201-210.