I find it hard to get ggplot for building polygons with holes. The following is shown. First get shapefile with git clone https://github.com/geotheory/volcano. Further:
require(ggplot2); require(ggmap); require(dplyr); require(maptools)
v = readShapePoly('volcano/volcano.shp')
v@proj4string = CRS('+proj=longlat +datum=WGS84')
plot(t(bbox(v)), type='l', lwd=8)
plot(v, col=paste0(colorRampPalette(c('grey','red'))(8),'dd'), add=T)

Looks nice. A ddalpha should display an invisible line if it is hidden by multiple polygons. Now try in ggplot.
d = fortify(v) %>% as_data_frame()
bb = bbox(v)
toner = get_stamenmap(c(bb[1,1], bb[2,1], bb[1,2], bb[2,2]), zoom=11, maptype='toner')
ggmap(toner) + geom_polygon(data=d, aes(long, lat, group=group, fill=id), alpha=.5)

The center polygons must overlap because the basemap is completely hidden in the center. Let me check the reinforced data for the holes:
d %>% select(id, hole) %>% table()
hole
id FALSE TRUE
0 278 0
1 715 0
2 392 388
3 388 331
4 390 265
5 265 387
6 328 125
7 125 0
It looks good, so try to visualize them individually.
i = 3
plot(v[i,], col='red')
ggplot(filter(d, id == i-1)) + geom_polygon(aes(long, lat, group=group, col=hole), fill=NA)
ggplot() + geom_polygon(data=filter(d, id==i-1), aes(long,lat, group=group))

Something seems wrong. Ggplot seems to ignore holes. If this is not a problem with the shapefile. Any suggestions on how to diagnose / fix this?