R help - Jun 2011 - nls Error Message - Singular Gradient Matrix

I've been trying to run some analysis using the nls function in R but keep 
coming up with an error message which I don't understand how to fix. The 
message follows here:

Error in nls(formula = f.p, data = n.data, start = list(S = 1, a = -0.1, : 

  singular gradient
In addition: Warning messages:
1: In min(x) : no non-missing arguments to min; returning Inf
2: In max(x) : no non-missing arguments to max; returning -Inf
> 

I did take a look at the archives before positing this and noticed that 
there was a similar message from a few years ago. The solution there 
seemed to be to change the algorithm to plinear. I've tried this to no 
avail. Unfortunately I'm not in a position where I can share the data
I'm
using, but will copy my code below. What I can say is that there are 
relatively few observations - only 12 - compared to the number of 
parameters. Could this be causing the error I'm seeing? Is there anything 
else that might be causing it? Any insight would be greatly appreciated!

Chris.

Code:

data <- read.csv(choose.files(),header=TRUE)
adoption <- data[,3]
year <- data[,2]
time <- data[,1]
cum.adoption <- data[,4]
plot(data[,2],data[,3],xlab="Year",ylab="Adoption",main="Greek
Cell Phone
Adoption 1994 - 2005 Q3",type="b")
plot(data[,2],data[,4],xlab="year",ylab="Cumulative
Adoption",main="Greek
Cell Phone Adoption 1994 - 2005 Q3",type="b")
start.val <- c(n=1.17,p=0.001,q=0.626)
bass <- nls(formula=data[,3] ~ p*n+(q-p)*data[,5]-(q/n)*data[,5]^2,
            data=n.data,
            start=start.val)
start.val <- c(S=1,a=1,b=1)
f.p <- cum.adoption~S/(1+exp(-a+b*time))
fisher.pry <- nls(formula = f.p,
                  data = n.data,
                  start= list(S=1,a=-0.1,b=0.1),
                  alg = "plinear",
                  trace=TRUE)
	[[alternative HTML version deleted]]