In a blog post in 2005, Ben Zimmer proposed a category of words called "acroblends" which share properties of both acronyms and portmaneaus. For example, Terracotta (for "Tool for Educational Research with RAndomized COnTrolled TriAls") is not strictly an acronym because it uses more than just the initial letter of each constituent word, but it's also not strictly a portmanteau, because "Terracotta" doesn't mean anything about experimental education research (whereas "smog" clearly means smoke+fog). Later, in a blog post in 2013, Arnold Zwicky proposed the word "acromanteau" to be a better label for this category of words. I agree, and as best I can tell, Zwicky is the first person to use the word "acromanteau" on the internet, although it was independently invented by Rick Hullinger in an email thread that motivated the current repository.
This repository contains a script for identifying words whose letters are ordered constituents of a larger set of words.
Here's the script: acromanteau.R
Usage:
# Clean-up
rm(list=ls())
# If you've downloaded the script to your local working directory:
# source('acromanteau.R')
# Or source directly from online GitHub repository:
source('https://raw.githubusercontent.com/bmotz/acromanteau/main/acromanteau.R')
# Words to Test
testwords <- c('random','word','alloy')
# Minimum length of matching acromanteaus
minlength <- 5
# Restrict order of test words?
restrict <- FALSE
output <- acromanteau(testwords,minlength,restrict)